Protótipo I.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Protótipo I - Sistema de recomendação baseado em modelos de <i>Machine Learning</i>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nesse notebook, teremos o desenvolvimento da primeira parte prática, do projeto relacionado ao TCC de tema e título acima citado."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:06.857341Z",
     "start_time": "2021-06-24T15:42:06.848339Z"
    }
   },
   "outputs": [],
   "source": [
    "# CONSTANTES\n",
    "\n",
    "PATH_TO_FULL_CF_FILE = \"preprocessed-data/CF/data_cf.pkl\"\n",
    "\n",
    "PATH_TO_MOVIES_CF_FILE = \"preprocessed-data/CF/movies_cf.pkl\"\n",
    "\n",
    "PATH_TO_RATINGS_CF_FILE = \"preprocessed-data/CF/ratings_cf.pkl\"\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "PATH_TO_FULL_CB_FILE = \"preprocessed-data/CB/data_cb.pkl\"\n",
    "\n",
    "PATH_TO_MOVIES_CB_FILE = \"preprocessed-data/CB/movies_cb.pkl\"\n",
    "\n",
    "PATH_TO_RATINGS_CB_FILE = \"preprocessed-data/CB/ratings_cb.pkl\"\n",
    "\n",
    "PATH_TO_RATINGS_INFOS_CB_FILE = \"preprocessed-data/CB/ratings_info_cb.pkl\"\n",
    "\n",
    "PATH_TO_TAG_RELEVANCE_GROUPED_CB_FILE = \"preprocessed-data/CB/tag_relevance_grouped_cb.pkl\"\n",
    "\n",
    "PATH_TO_TAG_RELEVANCE_CB_FILE = \"preprocessed-data/CB/tag_relevance_cb.pkl\"\n",
    "\n",
    "PATH_TO_TAGS_PROCESSED_CB_FILE = \"preprocessed-data/CB/tags_processed_cb.pkl\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:12.274162Z",
     "start_time": "2021-06-24T15:42:07.147026Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: imdbpy in c:\\users\\seaskythe\\anaconda3\\lib\\site-packages (2021.4.18)\n",
      "Requirement already satisfied: lxml in c:\\users\\seaskythe\\anaconda3\\lib\\site-packages (from imdbpy) (4.6.3)\n",
      "Requirement already satisfied: SQLAlchemy in c:\\users\\seaskythe\\anaconda3\\lib\\site-packages (from imdbpy) (1.4.29)\n",
      "Requirement already satisfied: greenlet!=0.4.17 in c:\\users\\seaskythe\\anaconda3\\lib\\site-packages (from SQLAlchemy->imdbpy) (1.1.2)\n"
     ]
    }
   ],
   "source": [
    "# Importando bibliotecas necessárias\n",
    "import pandas as pd\n",
    "pd.set_option(\"display.max_rows\", 25)\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "import sys\n",
    "import sys\n",
    "!{sys.executable} -m pip install imdbpy\n",
    "\n",
    "from imdb import IMDb\n",
    "\n",
    "# Importando garbage collector\n",
    "import gc\n",
    "\n",
    "# Importando bibliotecas para o sistema de recomendação\n",
    "from scipy.sparse import csr_matrix # Matriz esparsa\n",
    "\n",
    "from sklearn.neighbors import NearestNeighbors\n",
    "\n",
    "# Importando Regular Expression operations\n",
    "import re"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Um pouco sobre o dataset:\n",
    "\n",
    "O dataset tem 6 arquivos a serem carregados.\n",
    "\n",
    "O primeiro arquivo é o \"<i>ratings.csv</i>\".\n",
    "\n",
    "Nesse arquivo estão contidas todas as avaliações dadas a um filme (identificado pelo campo <b><i>movieID</i></b>), por um usuário (identificado pelo campo <b><i>userId</i></b>). \n",
    "\n",
    "A tabela tem as seguintes colunas: userId, movieId, rating, timestamp\".\n",
    "\n",
    "As notas dadas estão numa escala \"5 estrelas\", com incrementos de \"meia estrela\". (0.5 estrelas - 5.0 estrelas).\n",
    "\n",
    "Timestamp, que está contido na tabela representa segundos desde meia noite (UTC) de Janeiro de 1970.\n",
    "\n",
    "-------------------------------------------------------------------------------------------------------------------------------\n",
    "\n",
    "O segundo arquivo é o \"<i>movies.csv</i>\".\n",
    "\n",
    "Nesse arquivo estão contidas todas as informações sobre os filmes, para cada <b><i>movieID\\d</i></b> temos um <i>title</i> e <i>genres</i> atribuidos a ele.\n",
    "\n",
    "A tabela tem as seguintes colunas: movieId, title, genres.\n",
    "\n",
    "A lista de generos possíveis é a seguinte:\n",
    "<ul> \n",
    "    <li>Action</li>\n",
    "    <li>Adventure</li>\n",
    "    <li>Animation</li>\n",
    "    <li>Children's</li>\n",
    "    <li>Comedy</li>\n",
    "    <li>Crime</li>\n",
    "    <li>Documentary</li>\n",
    "    <li>Drama</li>\n",
    "    <li>Fantasy</li>\n",
    "    <li>Film-Noir</li>\n",
    "    <li>Horror</li>\n",
    "    <li>Musical</li>\n",
    "    <li>Mystery</li>\n",
    "    <li>Romance</li>\n",
    "    <li>Sci-Fi</li>\n",
    "    <li>Thriller</li>\n",
    "    <li>War</li>\n",
    "    <li>Western</li>\n",
    "    <li>(no genres listed)</li>\n",
    "</ul>\n",
    "\n",
    "-------------------------------------------------------------------------------------------------------------------------------\n",
    "-------------------------------------------------------------------------------------------------------------------------------\n",
    "\n",
    "O terceiro arquivo é o \"<i>tags.csv</i>\".\n",
    "\n",
    "Nesse arquivo estão contidas todas as tags aplicadas pelos usuários aos filmes, para cada <b><i>movieID</i></b> temos uma <i>tag</i> atribuida a ele.\n",
    "\n",
    "A tabela tem as seguintes colunas: userId, movieId, tag, timestamp.\n",
    "\n",
    "Tags são metadados gerados pelos usuários para cada filme. Cada tag é uma palavra ou frase curta.\n",
    "\n",
    "Timestamp, que está contido na tabela representa segundos desde meia noite (UTC) de Janeiro de 1970.\n",
    "\n",
    "-------------------------------------------------------------------------------------------------------------------------------\n",
    "-------------------------------------------------------------------------------------------------------------------------------\n",
    "\n",
    "O quarto arquivo é o \"<i>links.csv</i>\".\n",
    "\n",
    "Esse arquivo é ótimo para associar os dados que temos com outras fontes de dados, pois, cada linha tem os campos: movieId, imdbId, tmdbId. Ou seja, ele associa os filmes que temos com seus respectivos Ids em outras duas plataformas (imdb e themoviedb).\n",
    "\n",
    "-------------------------------------------------------------------------------------------------------------------------------\n",
    "-------------------------------------------------------------------------------------------------------------------------------\n",
    "\n",
    "Por ultimo temos o dito \"Tag Genome\" (\"<i>genome-scores.csv e genome-tags.csv</i>\".\n",
    "\n",
    "Tag Genome é uma estrutura de dados que contem pontuações de relevância para cada tag. É basicamente uma matriz (filmes x tags), onde cada filme no genoma tem um valor para cada tag. Essa matriz foi feita utilizando um algoritmo de machine learning sob dados obtidos de usuários, incluindo tags, avaliações e reviews textuais.\n",
    "\n",
    "\n",
    "Informações sobre o dataset retiradas de: https://files.grouplens.org/datasets/movielens/ml-25m-README.html"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1 - Preparação dos dados para Collaborative Filtering\n",
    "\n",
    "Nesse primeiro momento, iremos utilizar os dados sobre as avaliações e sobre os filmes para criar uma matriz filmes x usuário, com as notas dadas.\n",
    "\n",
    "Para resumir e relembrar, filtragem colaborativa leva em consideração a semelhança entre usuários ou entre itens para realizar uma recomendação, nada além disso. Por isso criaremos essa matriz de relação."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:19.659350Z",
     "start_time": "2021-06-24T15:42:12.275163Z"
    }
   },
   "outputs": [],
   "source": [
    "# Carregando arquivos CSV necessários para essa abordagem\n",
    "data_ratings = pd.read_csv(\"dataset/ml-25m/ratings.csv\")\n",
    "data_movies = pd.read_csv(\"dataset/ml-25m/movies.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:19.675353Z",
     "start_time": "2021-06-24T15:42:19.660349Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>userId</th>\n",
       "      <th>movieId</th>\n",
       "      <th>rating</th>\n",
       "      <th>timestamp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>296</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1147880044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>306</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1147868817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>307</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1147868828</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>665</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1147878820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>899</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1147868510</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   userId  movieId  rating   timestamp\n",
       "0       1      296     5.0  1147880044\n",
       "1       1      306     3.5  1147868817\n",
       "2       1      307     5.0  1147868828\n",
       "3       1      665     5.0  1147878820\n",
       "4       1      899     3.5  1147868510"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Exibindo registros inciais de ratings\n",
    "data_ratings.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:19.691357Z",
     "start_time": "2021-06-24T15:42:19.677354Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "25000095\n"
     ]
    }
   ],
   "source": [
    "# Verificando quantas avaliações temos\n",
    "print(data_ratings.shape[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:19.707361Z",
     "start_time": "2021-06-24T15:42:19.692358Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Toy Story (1995)</td>\n",
       "      <td>Adventure|Animation|Children|Comedy|Fantasy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>Jumanji (1995)</td>\n",
       "      <td>Adventure|Children|Fantasy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Grumpier Old Men (1995)</td>\n",
       "      <td>Comedy|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>Waiting to Exhale (1995)</td>\n",
       "      <td>Comedy|Drama|Romance</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>Father of the Bride Part II (1995)</td>\n",
       "      <td>Comedy</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId                               title  \\\n",
       "0        1                    Toy Story (1995)   \n",
       "1        2                      Jumanji (1995)   \n",
       "2        3             Grumpier Old Men (1995)   \n",
       "3        4            Waiting to Exhale (1995)   \n",
       "4        5  Father of the Bride Part II (1995)   \n",
       "\n",
       "                                        genres  \n",
       "0  Adventure|Animation|Children|Comedy|Fantasy  \n",
       "1                   Adventure|Children|Fantasy  \n",
       "2                               Comedy|Romance  \n",
       "3                         Comedy|Drama|Romance  \n",
       "4                                       Comedy  "
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Exibindo registros inciais de movies\n",
    "data_movies.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:19.915407Z",
     "start_time": "2021-06-24T15:42:19.708361Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "      <th>year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Toy Story</td>\n",
       "      <td>Adventure|Animation|Children|Comedy|Fantasy</td>\n",
       "      <td>1995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>Jumanji</td>\n",
       "      <td>Adventure|Children|Fantasy</td>\n",
       "      <td>1995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Grumpier Old Men</td>\n",
       "      <td>Comedy|Romance</td>\n",
       "      <td>1995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>Waiting to Exhale</td>\n",
       "      <td>Comedy|Drama|Romance</td>\n",
       "      <td>1995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>Father of the Bride Part II</td>\n",
       "      <td>Comedy</td>\n",
       "      <td>1995</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId                        title  \\\n",
       "0        1                    Toy Story   \n",
       "1        2                      Jumanji   \n",
       "2        3             Grumpier Old Men   \n",
       "3        4            Waiting to Exhale   \n",
       "4        5  Father of the Bride Part II   \n",
       "\n",
       "                                        genres  year  \n",
       "0  Adventure|Animation|Children|Comedy|Fantasy  1995  \n",
       "1                   Adventure|Children|Fantasy  1995  \n",
       "2                               Comedy|Romance  1995  \n",
       "3                         Comedy|Drama|Romance  1995  \n",
       "4                                       Comedy  1995  "
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Removendo os anos dos titulos dos filmes utilizando expressões regulares\n",
    "\n",
    "#Pegando os anos dos titulos e colocando numa nova coluna\n",
    "data_movies['year'] = data_movies.title.str.extract('(\\(\\d\\d\\d\\d\\))', expand = False)\n",
    "#Removendo os parênteses\n",
    "data_movies['year'] = data_movies.year.str.extract('(\\d\\d\\d\\d)', expand = False)\n",
    "#Removendo os anos da coluna de titulos\n",
    "data_movies['title'] = data_movies.title.str.replace('(\\(\\d\\d\\d\\d\\))', '', regex=True)\n",
    "#Garantindo que não tenha nenhum espaço sobrando\n",
    "data_movies['title'] = data_movies['title'].apply(lambda x: x.strip())\n",
    "\n",
    "data_movies.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:19.931411Z",
     "start_time": "2021-06-24T15:42:19.916408Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "movieId      0\n",
      "title        0\n",
      "genres       0\n",
      "year       410\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Verificando se algum valor ficou faltante\n",
    "print(data_movies.isnull().sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:20.011428Z",
     "start_time": "2021-06-24T15:42:19.933411Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "movieId    0\n",
       "title      0\n",
       "genres     0\n",
       "year       0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Percebemos que 410 filmes não tinham o ano registrado em seus títulos, vamos atribuir a esses filmes o valor 0 ao ano\n",
    "# e posteriormente podemos tratar melhor essas ocorrências\n",
    "\n",
    "# Primeiro vamos converter os anos para o tipo String, para facilitar a comparação\n",
    "data_movies[\"year\"] = data_movies[\"year\"].astype(str)\n",
    "\n",
    "# Aqui vamos percorrer as linhas e substituir os valores faltantes para 0\n",
    "for row in data_movies.itertuples():\n",
    "    if(row.year == 'nan'):\n",
    "        data_movies.at[row.Index, \"year\"] = 0\n",
    "        \n",
    "\n",
    "# Verificando novamente se existe algum valor nulo, após converter para int\n",
    "data_movies[\"year\"].astype(int)\n",
    "\n",
    "data_movies.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:20.027432Z",
     "start_time": "2021-06-24T15:42:20.012430Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "62423\n"
     ]
    }
   ],
   "source": [
    "# Verificando quantos filmes temos\n",
    "print(data_movies.shape[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:20.059439Z",
     "start_time": "2021-06-24T15:42:20.028433Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cinderella                 16\n",
      "Hamlet                      9\n",
      "Home                        9\n",
      "First Love                  9\n",
      "The Escape                  8\n",
      "                           ..\n",
      "Story of Ruth, The          1\n",
      "No Nukes                    1\n",
      "Fagbug                      1\n",
      "Attila                      1\n",
      "Women of Devil's Island     1\n",
      "Name: title, Length: 58529, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Verficando de fato quantos filmes temos\n",
    "print(data_movies[\"title\"].value_counts())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nota-se que alguns filmes estão repetidos, porém não é necessário removê-los agora, visto que esse problema irá se resolver durante o nosso processamento."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-26T19:36:37.055290Z",
     "start_time": "2021-05-26T19:36:37.040287Z"
    }
   },
   "source": [
    "#### Problemas observados: \n",
    "\n",
    "Como observado acima, temos alguns problemas, considerando uma abordagem Colaborativa, não precisamos das colunas \"timestamp\" presente em \"ratings\" e nem das colunas \"genres\" e \"year\" presente em \"movies\". Portanto vamos removê-las."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:20.387513Z",
     "start_time": "2021-06-24T15:42:20.060439Z"
    }
   },
   "outputs": [],
   "source": [
    "movies_cf = data_movies.loc[:, {\"movieId\", \"title\"}]\n",
    "ratings_cf = data_ratings.loc[:, {\"userId\", \"movieId\", \"rating\"}]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:20.435524Z",
     "start_time": "2021-06-24T15:42:20.388513Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Movies: \n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>title</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Toy Story</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>Jumanji</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Grumpier Old Men</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>Waiting to Exhale</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>Father of the Bride Part II</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId                        title\n",
       "0        1                    Toy Story\n",
       "1        2                      Jumanji\n",
       "2        3             Grumpier Old Men\n",
       "3        4            Waiting to Exhale\n",
       "4        5  Father of the Bride Part II"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Movies: \")\n",
    "# Salvando DataFrame de Filmes para collaborative filtering\n",
    "movies_cf.to_pickle(PATH_TO_MOVIES_CF_FILE)\n",
    "\n",
    "movies_cf.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:25.758308Z",
     "start_time": "2021-06-24T15:42:20.436525Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ratings: \n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>userId</th>\n",
       "      <th>rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>296</td>\n",
       "      <td>1</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>306</td>\n",
       "      <td>1</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>307</td>\n",
       "      <td>1</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>665</td>\n",
       "      <td>1</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>899</td>\n",
       "      <td>1</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId  userId  rating\n",
       "0      296       1     5.0\n",
       "1      306       1     3.5\n",
       "2      307       1     5.0\n",
       "3      665       1     5.0\n",
       "4      899       1     3.5"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(\"Ratings: \")\n",
    "# Salvando DataFrame de Ratings para collaborative filtering\n",
    "ratings_cf.to_pickle(PATH_TO_RATINGS_CF_FILE)\n",
    "\n",
    "ratings_cf.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Agora iremos juntar os dois dataframes e criar uma matriz usuário x filme."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:45.536495Z",
     "start_time": "2021-06-24T15:42:42.363028Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>title</th>\n",
       "      <th>userId</th>\n",
       "      <th>rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Toy Story</td>\n",
       "      <td>2</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Toy Story</td>\n",
       "      <td>3</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>Toy Story</td>\n",
       "      <td>4</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>Toy Story</td>\n",
       "      <td>5</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>Toy Story</td>\n",
       "      <td>8</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId      title  userId  rating\n",
       "0        1  Toy Story       2     3.5\n",
       "1        1  Toy Story       3     4.0\n",
       "2        1  Toy Story       4     3.0\n",
       "3        1  Toy Story       5     4.0\n",
       "4        1  Toy Story       8     4.0"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Juntando DataFrames\n",
    "data_cf = pd.merge(movies_cf, ratings_cf)\n",
    "data_cf.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:54.101026Z",
     "start_time": "2021-06-24T15:42:47.524566Z"
    }
   },
   "outputs": [],
   "source": [
    "# Salvando DataFrame para Collaborative Filtering\n",
    "data_cf.to_pickle(PATH_TO_FULL_CF_FILE)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nesse ponto temos uma matriz que pode ser utilizada, por exemplo, para o treinamento de um modelo que utiliza o KNN (K Nearest Neighbors) como algoritmo."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2 - Preparação dos dados para Content-Based"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:54.116030Z",
     "start_time": "2021-06-24T15:42:54.102026Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "      <th>year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Toy Story</td>\n",
       "      <td>Adventure|Animation|Children|Comedy|Fantasy</td>\n",
       "      <td>1995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>Jumanji</td>\n",
       "      <td>Adventure|Children|Fantasy</td>\n",
       "      <td>1995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Grumpier Old Men</td>\n",
       "      <td>Comedy|Romance</td>\n",
       "      <td>1995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>Waiting to Exhale</td>\n",
       "      <td>Comedy|Drama|Romance</td>\n",
       "      <td>1995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>Father of the Bride Part II</td>\n",
       "      <td>Comedy</td>\n",
       "      <td>1995</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId                        title  \\\n",
       "0        1                    Toy Story   \n",
       "1        2                      Jumanji   \n",
       "2        3             Grumpier Old Men   \n",
       "3        4            Waiting to Exhale   \n",
       "4        5  Father of the Bride Part II   \n",
       "\n",
       "                                        genres  year  \n",
       "0  Adventure|Animation|Children|Comedy|Fantasy  1995  \n",
       "1                   Adventure|Children|Fantasy  1995  \n",
       "2                               Comedy|Romance  1995  \n",
       "3                         Comedy|Drama|Romance  1995  \n",
       "4                                       Comedy  1995  "
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_movies.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:54.132035Z",
     "start_time": "2021-06-24T15:42:54.118029Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>userId</th>\n",
       "      <th>movieId</th>\n",
       "      <th>rating</th>\n",
       "      <th>timestamp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>296</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1147880044</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>306</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1147868817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>307</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1147868828</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>665</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1147878820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>899</td>\n",
       "      <td>3.5</td>\n",
       "      <td>1147868510</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   userId  movieId  rating   timestamp\n",
       "0       1      296     5.0  1147880044\n",
       "1       1      306     3.5  1147868817\n",
       "2       1      307     5.0  1147868828\n",
       "3       1      665     5.0  1147878820\n",
       "4       1      899     3.5  1147868510"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_ratings.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Na abordagem content-based, trabalhamos bastante com as semelhanças entre itens, no nosso caso filmes, no que diz respeito as suas características (ano, diretor, gênero, sinopse e etc).\n",
    "\n",
    "Dessa forma, a primeira coisa que podemos fazer é separar o título que está no formato Titulo (ANO), e construir duas colunas, uma título e uma com o ano. Para isso vamos utilizar expressões regulares."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:54.148037Z",
     "start_time": "2021-06-24T15:42:54.134033Z"
    }
   },
   "outputs": [],
   "source": [
    "# criando uma copia do DataFrame sobre os filmes\n",
    "movies_cb = data_movies.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:54.196047Z",
     "start_time": "2021-06-24T15:42:54.149037Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Figaro and Cleo                                                                               1\n",
       "City of the Living Dead (a.k.a. Gates of Hell, The) (Paura nella città dei morti viventi)     1\n",
       "Endless Summer, The                                                                           1\n",
       "Blow-Out (La grande bouffe)                                                                   1\n",
       "The Slipper and the Rose: The Story of Cinderella                                             1\n",
       "                                                                                             ..\n",
       "The Escape                                                                                    8\n",
       "First Love                                                                                    9\n",
       "Home                                                                                          9\n",
       "Hamlet                                                                                        9\n",
       "Cinderella                                                                                   16\n",
       "Name: title, Length: 58529, dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# verificando se existem filmes duplicados\n",
    "movies_cb[\"title\"].value_counts().sort_values()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nota-se que temos filmes duplicados no dataset, vamos então remove-los posteriormente"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "      <th>year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Toy Story</td>\n",
       "      <td>Adventure|Animation|Children|Comedy|Fantasy</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>Jumanji</td>\n",
       "      <td>Adventure|Children|Fantasy</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Grumpier Old Men</td>\n",
       "      <td>Comedy|Romance</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>Waiting to Exhale</td>\n",
       "      <td>Comedy|Drama|Romance</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>Father of the Bride Part II</td>\n",
       "      <td>Comedy</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId                        title  \\\n",
       "0        1                    Toy Story   \n",
       "1        2                      Jumanji   \n",
       "2        3             Grumpier Old Men   \n",
       "3        4            Waiting to Exhale   \n",
       "4        5  Father of the Bride Part II   \n",
       "\n",
       "                                        genres year  \n",
       "0  Adventure|Animation|Children|Comedy|Fantasy  NaN  \n",
       "1                   Adventure|Children|Fantasy  NaN  \n",
       "2                               Comedy|Romance  NaN  \n",
       "3                         Comedy|Drama|Romance  NaN  \n",
       "4                                       Comedy  NaN  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Removendo os anos dos titulos dos filmes utilizando expressões regulares\n",
    "\n",
    "#Pegando os anos dos titulos e colocando numa nova coluna\n",
    "data_movies['year'] = data_movies.title.str.extract('(\\(\\d\\d\\d\\d\\))', expand = False)\n",
    "#Removendo os parênteses\n",
    "data_movies['year'] = data_movies.year.str.extract('(\\d\\d\\d\\d)', expand = False)\n",
    "#Removendo os anos da coluna de titulos\n",
    "data_movies['title'] = data_movies.title.str.replace('(\\(\\d\\d\\d\\d\\))', '', regex=True)\n",
    "#Garantindo que não tenha nenhum espaço sobrando\n",
    "data_movies['title'] = data_movies['title'].apply(lambda x: x.strip())\n",
    "\n",
    "data_movies.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Modificando a forma com que os gêneros são apresentados\n",
    "\n",
    "Observando agora a coluna \"genres\", podemos perceber que o formato apresentado é um pouco complicado para uso, portanto vamos fazer algumas mudanças e conversões.\n",
    "\n",
    "Vamos transformar cada gênero numa coluna para melhorar isso, e se, o filme é daquele gênero, a coluna recebe o valor \"1\", caso contrario \"0\", dessa forma temos:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:54.287068Z",
     "start_time": "2021-06-24T15:42:54.197048Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>62413</th>\n",
       "      <td>209145</td>\n",
       "      <td>Liberté (2019)</td>\n",
       "      <td>[Drama]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62414</th>\n",
       "      <td>209147</td>\n",
       "      <td>The Carpet of Horror (1962)</td>\n",
       "      <td>[Crime, Horror]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62415</th>\n",
       "      <td>209151</td>\n",
       "      <td>Mao Zedong 1949 (2019)</td>\n",
       "      <td>[(no genres listed)]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62416</th>\n",
       "      <td>209153</td>\n",
       "      <td>Happy Flight (2008)</td>\n",
       "      <td>[Comedy, Drama]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62417</th>\n",
       "      <td>209155</td>\n",
       "      <td>Santosh Subramaniam (2008)</td>\n",
       "      <td>[Action, Comedy, Romance]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62418</th>\n",
       "      <td>209157</td>\n",
       "      <td>We (2018)</td>\n",
       "      <td>[Drama]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62419</th>\n",
       "      <td>209159</td>\n",
       "      <td>Window of the Soul (2001)</td>\n",
       "      <td>[Documentary]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62420</th>\n",
       "      <td>209163</td>\n",
       "      <td>Bad Poems (2018)</td>\n",
       "      <td>[Comedy, Drama]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62421</th>\n",
       "      <td>209169</td>\n",
       "      <td>A Girl Thing (2001)</td>\n",
       "      <td>[(no genres listed)]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62422</th>\n",
       "      <td>209171</td>\n",
       "      <td>Women of Devil's Island (1962)</td>\n",
       "      <td>[Action, Adventure, Drama]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       movieId                           title                      genres\n",
       "62413   209145                  Liberté (2019)                     [Drama]\n",
       "62414   209147     The Carpet of Horror (1962)             [Crime, Horror]\n",
       "62415   209151          Mao Zedong 1949 (2019)        [(no genres listed)]\n",
       "62416   209153             Happy Flight (2008)             [Comedy, Drama]\n",
       "62417   209155      Santosh Subramaniam (2008)   [Action, Comedy, Romance]\n",
       "62418   209157                       We (2018)                     [Drama]\n",
       "62419   209159       Window of the Soul (2001)               [Documentary]\n",
       "62420   209163                Bad Poems (2018)             [Comedy, Drama]\n",
       "62421   209169             A Girl Thing (2001)        [(no genres listed)]\n",
       "62422   209171  Women of Devil's Island (1962)  [Action, Adventure, Drama]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Substituindo os gêneros por listas\n",
    "movies_cb['genres'] = movies_cb['genres'].str.split('|')\n",
    "\n",
    "\n",
    "movies_cb.tail(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:56.003763Z",
     "start_time": "2021-06-24T15:42:55.978309Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "movieId\n",
       "1         1\n",
       "162130    1\n",
       "162070    1\n",
       "162072    1\n",
       "162074    1\n",
       "         ..\n",
       "107614    1\n",
       "107623    1\n",
       "107625    1\n",
       "107630    1\n",
       "209171    1\n",
       "Length: 62423, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "movies_cb.value_counts(\"movieId\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:56.643183Z",
     "start_time": "2021-06-24T15:42:56.521891Z"
    }
   },
   "outputs": [],
   "source": [
    "# Salvando DataFrame de Filmes para Content Based\n",
    "movies_cb.to_pickle(PATH_TO_MOVIES_CB_FILE)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Vamos agora juntar o DataFrame modificado com algumas informações que serão retiradas do Dataframe \"ratings\", porém antes, precisamos modificar o DataFrame \"ratings\" e remover seu campo \"timestamps\", calcular o rating médio para cada filme e verificar quantos votos cada filme tem."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:42:58.502814Z",
     "start_time": "2021-06-24T15:42:58.170005Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>userId</th>\n",
       "      <th>movieId</th>\n",
       "      <th>rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>296</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>306</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>307</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>665</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>899</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   userId  movieId  rating\n",
       "0       1      296     5.0\n",
       "1       1      306     3.5\n",
       "2       1      307     5.0\n",
       "3       1      665     5.0\n",
       "4       1      899     3.5"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Pegando apenas colunas relevantes de \"ratings\" - removemos timestamps\n",
    "ratings_cb = data_ratings.drop(\"timestamp\", axis='columns')\n",
    "ratings_cb.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:03.808691Z",
     "start_time": "2021-06-24T15:42:59.386001Z"
    }
   },
   "outputs": [],
   "source": [
    "# Salvando DataFrame de ratings para Content Based\n",
    "ratings_cb.to_pickle(PATH_TO_RATINGS_CB_FILE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:07.471447Z",
     "start_time": "2021-06-24T15:43:07.147376Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "36207        20\n",
       "31691        20\n",
       "135005       20\n",
       "135012       20\n",
       "51779        20\n",
       "          ...  \n",
       "20055      7488\n",
       "33844      7919\n",
       "137293     8913\n",
       "80974      9178\n",
       "72315     32202\n",
       "Name: userId, Length: 162541, dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Quantos usuarios participaram - 162541\n",
    "ratings_cb[\"userId\"].value_counts().sort_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:08.105590Z",
     "start_time": "2021-06-24T15:43:07.554467Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "200728        1\n",
       "188153        1\n",
       "194576        1\n",
       "182439        1\n",
       "178255        1\n",
       "          ...  \n",
       "2571      72674\n",
       "593       74127\n",
       "296       79672\n",
       "318       81482\n",
       "356       81491\n",
       "Name: movieId, Length: 59047, dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Contando quantos filmes temos no total - 59047\n",
    "ratings_cb[\"movieId\"].value_counts().sort_values()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nota-se que aqui temos menos que o valor inicial, isso se deve aos filmes repetidos na lista. (Porém o problema foi resolvido agora)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-16T17:17:07.043079Z",
     "start_time": "2021-06-16T17:17:07.032077Z"
    }
   },
   "source": [
    "A maioria dos filmes com nota máxima ou mínima tem pouquissimos votos (geralmente 1), e isso será mostrado mais pra frente. Porém, vamos pensar, caso estejamos procurando um restaurante e encontramos um restaurante com nota 5 com 1 voto, e um outro com 4.2 com 400 votos, qual escolheriamos? Obviamente escolheriamos o que tem 400 votos, por conta da confiabilidade.\n",
    "\n",
    "Como podemos então resolver esse problema? Utilizamentos o \"Weighted Rating\", uma métrica utilizada pelo IMDb para sobrepor esse problema, já que a quantidade de votos terá peso no calculo da nota.\n",
    "\n",
    "Vejamos como calcular esse rating."
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Percebendo que vários filmes tem nota máxima, ou mínima, percebemos que tais filmes são mais ou menos favorecidos caso o rating seja utilizado, para resolver isso, vamos utilizar o \"WeightedRating\" do IMDb ao invés do rating comum. A fórmula é a seguinte:\n",
    "\n",
    "![image.png](attachment:image.png)\n",
    "\n",
    "<ul> \n",
    "        <li>v: número de votos total do filme</li>\n",
    "        <li>m: número mínimo de votos para estar listado</li>\n",
    "        <li>R: rating médio do filme</li>\n",
    "        <li>C: rating médio total</li>\n",
    "</ul>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vamos então recuperar os dados necessários para fazer os calculos posteriormente"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "##### Pegando average rating de cada filme"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:14.915067Z",
     "start_time": "2021-06-24T15:43:13.850013Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>average rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>59042</th>\n",
       "      <td>209157</td>\n",
       "      <td>1.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59043</th>\n",
       "      <td>209159</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59044</th>\n",
       "      <td>209163</td>\n",
       "      <td>4.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59045</th>\n",
       "      <td>209169</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59046</th>\n",
       "      <td>209171</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       movieId  average rating\n",
       "59042   209157             1.5\n",
       "59043   209159             3.0\n",
       "59044   209163             4.5\n",
       "59045   209169             3.0\n",
       "59046   209171             3.0"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Agrupando pelo ID e fazendo a média do rating de cada filme - agrupamos pelo ID pois cada filme pode ter o mesmo titulo\n",
    "average_rating = pd.DataFrame(ratings_cb.groupby(\"movieId\").mean()[\"rating\"])\n",
    "\n",
    "# Voltando o index ao normal\n",
    "average_rating.reset_index(inplace=True)\n",
    "\n",
    "# Renomeando coluna\n",
    "average_rating.rename(columns={\"rating\": \"average rating\"}, inplace=True)\n",
    "\n",
    "average_rating.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T21:09:30.342793Z",
     "start_time": "2021-05-27T21:09:30.332791Z"
    }
   },
   "source": [
    "##### Pegando rating count de cada filme"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:17.107000Z",
     "start_time": "2021-06-24T15:43:16.194148Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>rating count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>59042</th>\n",
       "      <td>209157</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59043</th>\n",
       "      <td>209159</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59044</th>\n",
       "      <td>209163</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59045</th>\n",
       "      <td>209169</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59046</th>\n",
       "      <td>209171</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       movieId  rating count\n",
       "59042   209157             1\n",
       "59043   209159             1\n",
       "59044   209163             1\n",
       "59045   209169             1\n",
       "59046   209171             1"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Agrupando pelo ID contando quantos votos cada filme tem\n",
    "count_rating = pd.DataFrame(ratings_cb.groupby(\"movieId\").count()[\"rating\"])\n",
    "\n",
    "# Voltando o index ao normal\n",
    "count_rating.reset_index(inplace=True)\n",
    "\n",
    "# Renomeando coluna\n",
    "count_rating.rename(columns={\"rating\": \"rating count\"},inplace=True)\n",
    "\n",
    "count_rating.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Agora, efetivamente iremos juntar as informações dos Dataframes sobre ratings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:18.167203Z",
     "start_time": "2021-06-24T15:43:18.130195Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>average rating</th>\n",
       "      <th>rating count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>3.893708</td>\n",
       "      <td>57309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>3.251527</td>\n",
       "      <td>24228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>3.142028</td>\n",
       "      <td>11804</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>2.853547</td>\n",
       "      <td>2523</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>3.058434</td>\n",
       "      <td>11714</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId  average rating  rating count\n",
       "0        1        3.893708         57309\n",
       "1        2        3.251527         24228\n",
       "2        3        3.142028         11804\n",
       "3        4        2.853547          2523\n",
       "4        5        3.058434         11714"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Juntando DataFrames count_rating e average_rating\n",
    "rating_infos = pd.merge(average_rating, count_rating)\n",
    "rating_infos.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Aqui comprovamos a pouca quantidade de votos para filme com média maxima ou minima"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:19.325149Z",
     "start_time": "2021-06-24T15:43:19.306145Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>average rating</th>\n",
       "      <th>rating count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5693</th>\n",
       "      <td>5805</td>\n",
       "      <td>0.5</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9414</th>\n",
       "      <td>27914</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9640</th>\n",
       "      <td>31945</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15146</th>\n",
       "      <td>80154</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15290</th>\n",
       "      <td>80781</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58911</th>\n",
       "      <td>208597</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58963</th>\n",
       "      <td>208795</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58979</th>\n",
       "      <td>208859</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59033</th>\n",
       "      <td>209137</td>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59041</th>\n",
       "      <td>209155</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1398 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       movieId  average rating  rating count\n",
       "5693      5805             0.5             2\n",
       "9414     27914             5.0             1\n",
       "9640     31945             5.0             1\n",
       "15146    80154             0.5             1\n",
       "15290    80781             0.5             1\n",
       "...        ...             ...           ...\n",
       "58911   208597             5.0             1\n",
       "58963   208795             5.0             1\n",
       "58979   208859             5.0             1\n",
       "59033   209137             0.5             1\n",
       "59041   209155             5.0             1\n",
       "\n",
       "[1398 rows x 3 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Mostrando os filmes com nota média = 0.5 ou 5 \n",
    "rating_infos.loc[rating_infos['average rating'].isin([0.5, 5])]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Pegando dados e criando função de calculo de WR"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:22.039059Z",
     "start_time": "2021-06-24T15:43:22.026055Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Rating médio total:  3.071373920951152\n",
      "Valor de m:  413.0\n"
     ]
    }
   ],
   "source": [
    "# Calculando rating médio total\n",
    "C = rating_infos['average rating'].mean()\n",
    "\n",
    "print(\"Rating médio total: \", C)\n",
    "\n",
    "# Definindo numero de votos mínimo (25% mais votados)\n",
    "m = rating_infos['rating count'].quantile(0.90)\n",
    "print(\"Valor de m: \", m)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:22.402414Z",
     "start_time": "2021-06-24T15:43:22.386412Z"
    }
   },
   "outputs": [],
   "source": [
    "# Definindo a função que calcula o rating com peso. Parametros (series que indica a linha do filme, numero minimo de votos, rating medio total)\n",
    "def WeightedRating(movie_row, m=m, C=C):\n",
    "    v = movie_row['rating count']\n",
    "    R = movie_row['average rating']\n",
    "    \n",
    "    return ((v/(v+m)) * R) + ((m/(v+m)) * C)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:23.565851Z",
     "start_time": "2021-06-24T15:43:22.961715Z"
    }
   },
   "outputs": [],
   "source": [
    "# Criando a coluna que contem os ratings com peso\n",
    "# Para cada coluna ele vai pegar o numero de ratings e o médio, para fazer o calculo utizando a função acima\n",
    "rating_infos['weighted rating'] = rating_infos.apply(WeightedRating, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:24.564314Z",
     "start_time": "2021-06-24T15:43:24.522306Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>average rating</th>\n",
       "      <th>rating count</th>\n",
       "      <th>weighted rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>3.893708</td>\n",
       "      <td>57309</td>\n",
       "      <td>3.887824</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>3.251527</td>\n",
       "      <td>24228</td>\n",
       "      <td>3.248508</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>3.142028</td>\n",
       "      <td>11804</td>\n",
       "      <td>3.139640</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>3.058434</td>\n",
       "      <td>11714</td>\n",
       "      <td>3.058875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>3.854909</td>\n",
       "      <td>24588</td>\n",
       "      <td>3.841965</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24535</th>\n",
       "      <td>122886</td>\n",
       "      <td>3.739115</td>\n",
       "      <td>12678</td>\n",
       "      <td>3.718049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24543</th>\n",
       "      <td>122904</td>\n",
       "      <td>3.852662</td>\n",
       "      <td>13900</td>\n",
       "      <td>3.830118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28451</th>\n",
       "      <td>134130</td>\n",
       "      <td>4.033416</td>\n",
       "      <td>16489</td>\n",
       "      <td>4.009909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28743</th>\n",
       "      <td>134853</td>\n",
       "      <td>3.932401</td>\n",
       "      <td>13580</td>\n",
       "      <td>3.906988</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40051</th>\n",
       "      <td>164179</td>\n",
       "      <td>4.027511</td>\n",
       "      <td>10087</td>\n",
       "      <td>3.989903</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>580 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       movieId  average rating  rating count  weighted rating\n",
       "0            1        3.893708         57309         3.887824\n",
       "1            2        3.251527         24228         3.248508\n",
       "2            3        3.142028         11804         3.139640\n",
       "4            5        3.058434         11714         3.058875\n",
       "5            6        3.854909         24588         3.841965\n",
       "...        ...             ...           ...              ...\n",
       "24535   122886        3.739115         12678         3.718049\n",
       "24543   122904        3.852662         13900         3.830118\n",
       "28451   134130        4.033416         16489         4.009909\n",
       "28743   134853        3.932401         13580         3.906988\n",
       "40051   164179        4.027511         10087         3.989903\n",
       "\n",
       "[580 rows x 4 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Exibindo alguns registros, que agora contém os ratings com peso\n",
    "rating_infos.loc[(rating_infos['rating count'] > 10000) & (rating_infos['rating count'] < 60000)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:24.993025Z",
     "start_time": "2021-06-24T15:43:24.954017Z"
    }
   },
   "outputs": [],
   "source": [
    "# Salvando o rating infos para Content Based\n",
    "rating_infos.to_pickle(PATH_TO_RATINGS_INFOS_CB_FILE)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "####  Agora, jutando as informações dos filmes com as sobre ratings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:26.874991Z",
     "start_time": "2021-06-24T15:43:26.825980Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "      <th>year</th>\n",
       "      <th>average rating</th>\n",
       "      <th>rating count</th>\n",
       "      <th>weighted rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Toy Story</td>\n",
       "      <td>[Adventure, Animation, Children, Comedy, Fantasy]</td>\n",
       "      <td>1995</td>\n",
       "      <td>3.893708</td>\n",
       "      <td>57309</td>\n",
       "      <td>3.887824</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>Jumanji</td>\n",
       "      <td>[Adventure, Children, Fantasy]</td>\n",
       "      <td>1995</td>\n",
       "      <td>3.251527</td>\n",
       "      <td>24228</td>\n",
       "      <td>3.248508</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Grumpier Old Men</td>\n",
       "      <td>[Comedy, Romance]</td>\n",
       "      <td>1995</td>\n",
       "      <td>3.142028</td>\n",
       "      <td>11804</td>\n",
       "      <td>3.139640</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>Waiting to Exhale</td>\n",
       "      <td>[Comedy, Drama, Romance]</td>\n",
       "      <td>1995</td>\n",
       "      <td>2.853547</td>\n",
       "      <td>2523</td>\n",
       "      <td>2.884188</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>Father of the Bride Part II</td>\n",
       "      <td>[Comedy]</td>\n",
       "      <td>1995</td>\n",
       "      <td>3.058434</td>\n",
       "      <td>11714</td>\n",
       "      <td>3.058875</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId                        title  \\\n",
       "0        1                    Toy Story   \n",
       "1        2                      Jumanji   \n",
       "2        3             Grumpier Old Men   \n",
       "3        4            Waiting to Exhale   \n",
       "4        5  Father of the Bride Part II   \n",
       "\n",
       "                                              genres  year  average rating  \\\n",
       "0  [Adventure, Animation, Children, Comedy, Fantasy]  1995        3.893708   \n",
       "1                     [Adventure, Children, Fantasy]  1995        3.251527   \n",
       "2                                  [Comedy, Romance]  1995        3.142028   \n",
       "3                           [Comedy, Drama, Romance]  1995        2.853547   \n",
       "4                                           [Comedy]  1995        3.058434   \n",
       "\n",
       "   rating count  weighted rating  \n",
       "0         57309         3.887824  \n",
       "1         24228         3.248508  \n",
       "2         11804         3.139640  \n",
       "3          2523         2.884188  \n",
       "4         11714         3.058875  "
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Fazendo copia\n",
    "data_cb = movies_cb.copy()\n",
    "\n",
    "# Juntando dados dos filmes com as informações sobre rating\n",
    "data_cb = data_cb.merge(rating_infos)\n",
    "\n",
    "data_cb.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:29.039405Z",
     "start_time": "2021-06-24T15:43:29.002397Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>average rating</th>\n",
       "      <th>rating count</th>\n",
       "      <th>weighted rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>59047.000000</td>\n",
       "      <td>59047.000000</td>\n",
       "      <td>59047.000000</td>\n",
       "      <td>59047.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>120692.748065</td>\n",
       "      <td>3.071374</td>\n",
       "      <td>423.393144</td>\n",
       "      <td>3.099534</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>64452.508194</td>\n",
       "      <td>0.739840</td>\n",
       "      <td>2477.885821</td>\n",
       "      <td>0.149322</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.710100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>78277.000000</td>\n",
       "      <td>2.687500</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.067253</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>136782.000000</td>\n",
       "      <td>3.150000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>3.072235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>173812.000000</td>\n",
       "      <td>3.500000</td>\n",
       "      <td>36.000000</td>\n",
       "      <td>3.080515</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>209171.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>81491.000000</td>\n",
       "      <td>4.406807</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             movieId  average rating  rating count  weighted rating\n",
       "count   59047.000000    59047.000000  59047.000000     59047.000000\n",
       "mean   120692.748065        3.071374    423.393144         3.099534\n",
       "std     64452.508194        0.739840   2477.885821         0.149322\n",
       "min         1.000000        0.500000      1.000000         1.710100\n",
       "25%     78277.000000        2.687500      2.000000         3.067253\n",
       "50%    136782.000000        3.150000      6.000000         3.072235\n",
       "75%    173812.000000        3.500000     36.000000         3.080515\n",
       "max    209171.000000        5.000000  81491.000000         4.406807"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Verificando dados estatísticos gerais sobre o dataFrame\n",
    "data_cb.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Iremos importar agora tags.csv, genome-scores.csv, genome-tags.csv"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Importando tags.csv, genome-scores.csv e genome-tags.csv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:30.446004Z",
     "start_time": "2021-06-24T15:43:29.858996Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>userId</th>\n",
       "      <th>movieId</th>\n",
       "      <th>tag</th>\n",
       "      <th>timestamp</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>260</td>\n",
       "      <td>classic</td>\n",
       "      <td>1439472355</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>260</td>\n",
       "      <td>sci-fi</td>\n",
       "      <td>1439472256</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>1732</td>\n",
       "      <td>dark comedy</td>\n",
       "      <td>1573943598</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1732</td>\n",
       "      <td>great dialogue</td>\n",
       "      <td>1573943604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>7569</td>\n",
       "      <td>so bad it's good</td>\n",
       "      <td>1573943455</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   userId  movieId               tag   timestamp\n",
       "0       3      260           classic  1439472355\n",
       "1       3      260            sci-fi  1439472256\n",
       "2       4     1732       dark comedy  1573943598\n",
       "3       4     1732    great dialogue  1573943604\n",
       "4       4     7569  so bad it's good  1573943455"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Importando tags.csv\n",
    "data_tags = pd.read_csv(\"dataset/ml-25m/tags.csv\")\n",
    "data_tags.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:34.863355Z",
     "start_time": "2021-06-24T15:43:30.448004Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Genome-Scores: \n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>tagId</th>\n",
       "      <th>relevance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.02875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0.02375</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0.06250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>0.07575</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>0.14075</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId  tagId  relevance\n",
       "0        1      1    0.02875\n",
       "1        1      2    0.02375\n",
       "2        1      3    0.06250\n",
       "3        1      4    0.07575\n",
       "4        1      5    0.14075"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Importando genome-scores.csv\n",
    "data_genomeScores = pd.read_csv(\"dataset/ml-25m/genome-scores.csv\")\n",
    "\n",
    "print(\"Genome-Scores: \")\n",
    "data_genomeScores.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:34.911366Z",
     "start_time": "2021-06-24T15:43:34.864355Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Genome-Tags: \n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tagId</th>\n",
       "      <th>tag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>007 (series)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>18th century</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1920s</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>1930s</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   tagId           tag\n",
       "0      1           007\n",
       "1      2  007 (series)\n",
       "2      3  18th century\n",
       "3      4         1920s\n",
       "4      5         1930s"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Importando genome-tags.csv\n",
    "data_genomeTags = pd.read_csv(\"dataset/ml-25m/genome-tags.csv\")\n",
    "\n",
    "print(\"Genome-Tags: \")\n",
    "data_genomeTags.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Observando os DataFrames restantes, o primeiro passo é remover a coluna timestamps do DataFrame tags."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:34.927369Z",
     "start_time": "2021-06-24T15:43:34.912366Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>userId</th>\n",
       "      <th>movieId</th>\n",
       "      <th>tag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>260</td>\n",
       "      <td>classic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>260</td>\n",
       "      <td>sci-fi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4</td>\n",
       "      <td>1732</td>\n",
       "      <td>dark comedy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1732</td>\n",
       "      <td>great dialogue</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>7569</td>\n",
       "      <td>so bad it's good</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   userId  movieId               tag\n",
       "0       3      260           classic\n",
       "1       3      260            sci-fi\n",
       "2       4     1732       dark comedy\n",
       "3       4     1732    great dialogue\n",
       "4       4     7569  so bad it's good"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Removendo coluna timestamp\n",
    "data_tags.pop(\"timestamp\")\n",
    "data_tags.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### O primeiro trabalho a ser feito vai ser com o dataframe feito a partir de tags.csv"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Aqui o trabalho é um pouco mais criativo, pois as \"tags\" são criadas pelos usuários e, algumas delas representam mais opiniões do que outras. Então iremos definir quais iremos manter. Antes desse processo, vamos deixar todas as tags em letras minústucas e remover os parênteses de todas elas."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:41.661249Z",
     "start_time": "2021-06-24T15:43:41.226406Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['classic',\n",
       " 'sci-fi',\n",
       " 'dark comedy',\n",
       " 'great dialogue',\n",
       " \"so bad it's good\",\n",
       " 'unreliable narrators',\n",
       " 'tense',\n",
       " 'artificial intelligence',\n",
       " 'philosophical',\n",
       " 'cliche',\n",
       " 'musical',\n",
       " 'horror',\n",
       " 'unpredictable',\n",
       " 'oscar (best supporting actress)',\n",
       " 'adventure',\n",
       " 'anime',\n",
       " 'ecology',\n",
       " 'fantasy',\n",
       " 'hayao miyazaki',\n",
       " 'miyazaki',\n",
       " 'post-apocalyptic',\n",
       " 'bah',\n",
       " 'clint eastwood',\n",
       " 'music',\n",
       " 'art',\n",
       " 'contemporary art',\n",
       " 'documentary',\n",
       " 'aliens',\n",
       " 'amazing photography',\n",
       " 'director: james cameron',\n",
       " 'first contact',\n",
       " 'james cameron',\n",
       " 'michael biehn',\n",
       " 'android(s)/cyborg(s)',\n",
       " 'apocalypse',\n",
       " 'arnold schwarzenegger',\n",
       " 'franchise',\n",
       " 'terminator',\n",
       " 'time travel',\n",
       " 'science fiction',\n",
       " 'inferior sequel',\n",
       " 'setting:london (uk) (future)',\n",
       " 'unoriginal',\n",
       " 'bad science',\n",
       " 'good science',\n",
       " 'hans zimmer',\n",
       " 'philosophical issues',\n",
       " 'space',\n",
       " 'space travel',\n",
       " 'time-travel',\n",
       " 'visually appealing',\n",
       " 'bittersweet',\n",
       " 'boring',\n",
       " 'conversation',\n",
       " 'dialogue driven',\n",
       " 'loneliness',\n",
       " 'love story',\n",
       " 'philosophy',\n",
       " 'quirky',\n",
       " 'reflective',\n",
       " 'romantic',\n",
       " 'thought-provoking',\n",
       " 'vienna',\n",
       " 'wistful',\n",
       " 'american civil war',\n",
       " 'based on book',\n",
       " 'culture clash',\n",
       " 'drama',\n",
       " 'indians',\n",
       " 'war',\n",
       " 'western',\n",
       " 'buddhism',\n",
       " 'china',\n",
       " 'great cinematography',\n",
       " 'journey of self-discovery',\n",
       " 'mountains',\n",
       " 'traveling',\n",
       " 'atmospheric',\n",
       " 'based on a book',\n",
       " 'canada',\n",
       " 'death',\n",
       " 'incest',\n",
       " 'meditative',\n",
       " 'melancholic',\n",
       " 'nudity (full frontal)',\n",
       " 'small town',\n",
       " 'wintry',\n",
       " 'aging',\n",
       " 'based on a play',\n",
       " 'friendship',\n",
       " 'morgan freeman',\n",
       " 'no action',\n",
       " 'old age',\n",
       " 'racism',\n",
       " 'southern theme',\n",
       " 'usa',\n",
       " 'emotional',\n",
       " 'heartwarming',\n",
       " 'holocaust',\n",
       " 'italy',\n",
       " 'love',\n",
       " 'oscar (best foreign language film)',\n",
       " 'oscar winner',\n",
       " 'sentimental',\n",
       " 'survival',\n",
       " 'natalie portman',\n",
       " 'robots',\n",
       " 'star wars',\n",
       " 'abortion',\n",
       " 'charlize theron',\n",
       " 'medicine',\n",
       " 'orphans',\n",
       " 'tobey maguire',\n",
       " 'break-up',\n",
       " 'catherine zeta-jones',\n",
       " 'funny',\n",
       " 'relationships',\n",
       " 'island',\n",
       " 'dogs',\n",
       " 'multiple storylines',\n",
       " 'action',\n",
       " 'george lucas',\n",
       " 'romance',\n",
       " 'family',\n",
       " 'mystery',\n",
       " 'powerful ending',\n",
       " 'twist ending',\n",
       " 'adapted from:book',\n",
       " 'assassin',\n",
       " 'assassination',\n",
       " 'history',\n",
       " 'interesting',\n",
       " 'southern europe',\n",
       " 'thriller',\n",
       " 'dialogue',\n",
       " 'paris',\n",
       " 'adultery',\n",
       " 'biographical',\n",
       " 'biography',\n",
       " 'blindness',\n",
       " 'disability',\n",
       " 'drug addiction',\n",
       " 'drugs',\n",
       " 'genius',\n",
       " 'great acting',\n",
       " 'music business',\n",
       " 'rhythm & blues',\n",
       " 'catholicism',\n",
       " 'child abuse',\n",
       " 'enigmatic',\n",
       " 'gay sex',\n",
       " 'mistaken identity',\n",
       " 'queer',\n",
       " 'soccer',\n",
       " 'spanish',\n",
       " 'alcoholism',\n",
       " 'complex characters',\n",
       " 'confusing',\n",
       " 'germany',\n",
       " 'biomedical ethics',\n",
       " 'euthanasia',\n",
       " 'fighting the system',\n",
       " 'suicide',\n",
       " 'suicide attempt',\n",
       " 'terminal illness',\n",
       " 'touching',\n",
       " 'true story',\n",
       " 'deliberate',\n",
       " 'nonsense',\n",
       " 'philosophycal',\n",
       " 'quiet',\n",
       " 'silent',\n",
       " 'very little dialogue',\n",
       " 'biopic',\n",
       " 'character study',\n",
       " 'death penalty',\n",
       " 'gay',\n",
       " 'journalism',\n",
       " 'prison',\n",
       " 'thought provoking',\n",
       " 'writers',\n",
       " 'based on a true story',\n",
       " 'joaquin phoenix',\n",
       " 'disturbing',\n",
       " 'murder mystery',\n",
       " 'non-hollywood ending',\n",
       " 'twist',\n",
       " 'simple',\n",
       " 'straight forward',\n",
       " 'strong woman',\n",
       " 'women during wartime',\n",
       " 'world war ii',\n",
       " 'cynical',\n",
       " 'nudity (topless - notable)',\n",
       " 'nudity (topless)',\n",
       " 'psychology',\n",
       " 'epidemic',\n",
       " 'husband-wife relationship',\n",
       " 'science',\n",
       " 'scientific research',\n",
       " 'george clooney',\n",
       " 'lawyers',\n",
       " 'depressing',\n",
       " 'hospital',\n",
       " 'marriage',\n",
       " 'jude law',\n",
       " 'road movie',\n",
       " 'soundtrack',\n",
       " 'alternate endings',\n",
       " 'beautiful',\n",
       " 'humanity',\n",
       " 'quirky style',\n",
       " 'solitude',\n",
       " 'awkward',\n",
       " 'barcelona',\n",
       " 'threesome',\n",
       " 'congressman',\n",
       " 'corruption',\n",
       " 'mistress',\n",
       " 'murder',\n",
       " 'newspaper',\n",
       " 'political corruption',\n",
       " 'politics',\n",
       " 'secrets',\n",
       " 'humorous',\n",
       " 'nonlinear',\n",
       " 'excellent script',\n",
       " 'nostalgic',\n",
       " 'revenge',\n",
       " 'effects',\n",
       " 'environmental',\n",
       " 'alternate reality',\n",
       " 'dystopia',\n",
       " 'oscar (best picture)',\n",
       " 'mindfuck',\n",
       " \"alzheimer's disease\",\n",
       " 'animals',\n",
       " 'genetics',\n",
       " 'virus',\n",
       " 'children',\n",
       " 'cinematography',\n",
       " 'campaign manager',\n",
       " 'real life',\n",
       " 'fish',\n",
       " 'food',\n",
       " 'ginza',\n",
       " 'japan',\n",
       " 'michelin',\n",
       " 'sushi',\n",
       " 'tokyo',\n",
       " 'tsukiji',\n",
       " 'humor',\n",
       " 'sexuality',\n",
       " 'static',\n",
       " 'strange',\n",
       " 'uncomfortable',\n",
       " 'oscar (best directing)',\n",
       " 'us president',\n",
       " 'president; butler',\n",
       " 'ferrari',\n",
       " 'formula 1',\n",
       " 'great skill',\n",
       " 'interesting characters',\n",
       " 'racing',\n",
       " 'sports',\n",
       " 'happy ending',\n",
       " 'middle-aged protagonist',\n",
       " 'warm',\n",
       " 'captain',\n",
       " 'ocean',\n",
       " 'suspense',\n",
       " 'disease',\n",
       " 'matthew mcconaughey',\n",
       " 'medical',\n",
       " 'dramatic comedy',\n",
       " 'sad',\n",
       " 'story',\n",
       " 'narration',\n",
       " 'cats',\n",
       " 'folk music',\n",
       " 'new york city',\n",
       " 'repetitive',\n",
       " 'wall street',\n",
       " 'human computer interaction',\n",
       " 'landscape',\n",
       " 'letters',\n",
       " 'meaning of life',\n",
       " 'original plot',\n",
       " 'repeating',\n",
       " 'afghanistan',\n",
       " 'seals',\n",
       " 'naration',\n",
       " 'nudity (full frontal - notable)',\n",
       " 'german sync',\n",
       " 'matt damon',\n",
       " 'lars von trier',\n",
       " 'pornography',\n",
       " 'unexpected ending',\n",
       " 'interesting ideea',\n",
       " 'original',\n",
       " 'spectacle',\n",
       " 'wormhole',\n",
       " 'espionage',\n",
       " 'jean dujardin',\n",
       " 'bitterswee',\n",
       " 'new york',\n",
       " 'future',\n",
       " 'time loop',\n",
       " 'tom cruise',\n",
       " 'plot holes',\n",
       " 'ridiculous dialogue',\n",
       " 'transformers',\n",
       " 'weak story',\n",
       " 'education',\n",
       " 'good acting',\n",
       " 'jazz',\n",
       " 'motivation',\n",
       " 'perfectionism',\n",
       " 'psychological',\n",
       " 'ben affleck',\n",
       " 'false accusation',\n",
       " 'good adaptation',\n",
       " 'infidelity',\n",
       " 'meticulous',\n",
       " 'psychopathy',\n",
       " 'matt reeves',\n",
       " 'comic',\n",
       " 'superheroes',\n",
       " 'talking animals',\n",
       " 'vin diesel',\n",
       " 'counterespionage',\n",
       " 'slow',\n",
       " 'law',\n",
       " 'acting',\n",
       " 'brad pitt',\n",
       " 'david ayer',\n",
       " 'to much war',\n",
       " 'war hero',\n",
       " 'manipulations',\n",
       " 'technology',\n",
       " 'mental illness',\n",
       " 'surprising',\n",
       " 'twisted end',\n",
       " 'code breaking',\n",
       " 'computers',\n",
       " 'cryptography',\n",
       " 'informatics',\n",
       " 'mathematics',\n",
       " 'physics',\n",
       " 'stephen hawking',\n",
       " 'image',\n",
       " 'olga kurylenko',\n",
       " 'improvisation',\n",
       " 'mars',\n",
       " 'nasa',\n",
       " 'near future',\n",
       " 'predictible',\n",
       " 'scientific',\n",
       " 'survival instinct',\n",
       " 'adoption',\n",
       " 'oscar nominee: best picture',\n",
       " 'dreamlike',\n",
       " 'imagination',\n",
       " 'surreal',\n",
       " 'surrealism',\n",
       " 'truth vs. fantasy',\n",
       " 'gay romance',\n",
       " 'space adventure',\n",
       " 'overcoming obstacles',\n",
       " 'teamwork',\n",
       " 'coen brothers',\n",
       " 'witty',\n",
       " 'smart',\n",
       " 'tom hanks',\n",
       " '4th wall',\n",
       " 'crude language',\n",
       " 'parody',\n",
       " 'soulless',\n",
       " 'cult film',\n",
       " 'unintentional comedy',\n",
       " 'dark',\n",
       " 'missing children',\n",
       " 'torture',\n",
       " 'ewan mcgregor',\n",
       " 'jar jar binks',\n",
       " 'ai',\n",
       " 'karate',\n",
       " 'inspiring',\n",
       " 'purpose in life',\n",
       " 'summer',\n",
       " 'ambush',\n",
       " 'england',\n",
       " 'romans',\n",
       " 'violent',\n",
       " 'television',\n",
       " 'michael moore',\n",
       " 'propaganda',\n",
       " 'metafiction',\n",
       " 'christopher nolan',\n",
       " 'turing test',\n",
       " 'impressive visuals',\n",
       " 'magic',\n",
       " 'tang dynasty',\n",
       " 'nicolas cage',\n",
       " 'manipulation',\n",
       " 'free running',\n",
       " 'no story',\n",
       " 'parkour',\n",
       " 'running',\n",
       " 'abduction',\n",
       " 'bound',\n",
       " 'dark humor',\n",
       " 'gagged',\n",
       " 'kidnapping',\n",
       " 'montreal',\n",
       " 'nightmares',\n",
       " 'small cast',\n",
       " 'twisted',\n",
       " 'bad ending',\n",
       " 'nudity (topless - brief)',\n",
       " 'oscar (best actor)',\n",
       " 'oscar (best actress)',\n",
       " 'want to see again',\n",
       " 'sick',\n",
       " 'animation',\n",
       " 'big budget',\n",
       " 'christian',\n",
       " 'christmas',\n",
       " 'starts well but fails you in the middle till the end',\n",
       " 'film noir',\n",
       " 'low budget',\n",
       " 'decline of civilization',\n",
       " 'prophetic',\n",
       " 'sex with the enemy',\n",
       " 'brutal violence',\n",
       " 'gun fu',\n",
       " 'plot twist',\n",
       " 'stylized violence',\n",
       " 'too much violence',\n",
       " 'amazing!',\n",
       " 'anti-hero',\n",
       " 'better than expected',\n",
       " 'innovative',\n",
       " 'movie to see',\n",
       " 'superhero',\n",
       " 'will smith',\n",
       " 'enjoyable',\n",
       " 'good laughs',\n",
       " 'chick flick',\n",
       " 'space program',\n",
       " 'stephan king',\n",
       " 'special effects',\n",
       " 'off-beat comedy',\n",
       " 'disney',\n",
       " 'netflixq',\n",
       " 'cult classic',\n",
       " 'homoerotic subtext',\n",
       " 'anti-romantic ending',\n",
       " 'racist',\n",
       " 'great soundtrack',\n",
       " 'stylized',\n",
       " 'weird',\n",
       " 'studio ghibli',\n",
       " 'black comedy',\n",
       " 'comedy',\n",
       " 'awesome',\n",
       " 'colorful',\n",
       " 'jeff goldblum',\n",
       " 'taika waititi',\n",
       " 'unconventional',\n",
       " 'heartbreaking',\n",
       " 'classic sci-fi',\n",
       " 'good vs evil',\n",
       " 'brilliant',\n",
       " 'scary',\n",
       " 'ghosts',\n",
       " 'bisexual',\n",
       " 'canadian',\n",
       " 'distopian',\n",
       " 'indy',\n",
       " 'lgbt',\n",
       " 'thomas gibson',\n",
       " 'oldie but goodie',\n",
       " 'scifi cult',\n",
       " 'quentin tarantino',\n",
       " 'violence',\n",
       " 'courtroom',\n",
       " 'courtroom drama',\n",
       " 'group psychology',\n",
       " 'old',\n",
       " 'social commentary',\n",
       " 'college',\n",
       " 'geeks',\n",
       " 'nerds',\n",
       " 'smart comedy',\n",
       " 'teen comedy',\n",
       " 'dystopic future',\n",
       " 'futuristic',\n",
       " 'genetic selection',\n",
       " 'intelligent',\n",
       " 'uma thurman',\n",
       " 'underdog',\n",
       " 'jim carrey',\n",
       " 'poignant',\n",
       " 'reality tv',\n",
       " 'bisexuality',\n",
       " 'historic',\n",
       " 'sexy',\n",
       " 'conspiracy',\n",
       " 'cover up',\n",
       " 'political',\n",
       " 'satire',\n",
       " 'unique',\n",
       " 'crude humor',\n",
       " 'sexual',\n",
       " 'stupid',\n",
       " 'coming of age',\n",
       " 'coming-of-age',\n",
       " 'great music',\n",
       " 'john hughes',\n",
       " 'overdramatic',\n",
       " 'school drama',\n",
       " 'teen',\n",
       " 'teen movie',\n",
       " 'bromance',\n",
       " 'high school',\n",
       " 'reese witherspoon',\n",
       " 'stereotypes',\n",
       " 'menage a trois',\n",
       " 'polygamy',\n",
       " 'guilty pleasure',\n",
       " '1980s',\n",
       " 'breaking the fourth wall',\n",
       " 'charming',\n",
       " 'rebellion',\n",
       " 'complicated',\n",
       " 'funny moments',\n",
       " 'mean spirited',\n",
       " 'tolkien',\n",
       " 'cloning',\n",
       " 'immortality',\n",
       " 'hilarious',\n",
       " 'sex',\n",
       " 'sex comedy',\n",
       " 'steve carell',\n",
       " 'jonah hill',\n",
       " 'marijuana',\n",
       " 'quotable',\n",
       " 'stoner',\n",
       " 'stoner movie',\n",
       " 'video game testers',\n",
       " 'video games',\n",
       " 'underrated',\n",
       " 'emma stone',\n",
       " 'high school life',\n",
       " 'amy poehler',\n",
       " 'teacher',\n",
       " 'hawaii',\n",
       " 'jason segel',\n",
       " 'kristen bell',\n",
       " 'mila kunis',\n",
       " 'paul rudd',\n",
       " 'cia',\n",
       " 'father daughter relationship',\n",
       " 'gritty',\n",
       " 'human trafficking',\n",
       " 'buddy movie',\n",
       " 'silly',\n",
       " 'spoof',\n",
       " 'subtle humor',\n",
       " 'will ferrell',\n",
       " 'awesome soundtrack',\n",
       " 'cultural references',\n",
       " 'geek',\n",
       " 'geeky',\n",
       " 'protagonist is a teacher',\n",
       " 'jason bateman',\n",
       " 'jennifer aniston',\n",
       " 'weak ending',\n",
       " 'channing tatum',\n",
       " 'self-aware',\n",
       " 'undercover cops',\n",
       " 'election',\n",
       " 'brother sister incest',\n",
       " 'teacher student sex',\n",
       " 'cobie smulders',\n",
       " 'rom com',\n",
       " 'self deprecating',\n",
       " 'absurd',\n",
       " 'cross generational',\n",
       " 'travel',\n",
       " 'workplace humor',\n",
       " 'kevin hart',\n",
       " 'reverse racism',\n",
       " 'will ferrel',\n",
       " 'feminism',\n",
       " 'romantic comedy',\n",
       " 'overcomplicated',\n",
       " 'woody harrelson',\n",
       " 'uplifting',\n",
       " 'outsider',\n",
       " 'sibling relationship',\n",
       " 'teen angst',\n",
       " 'bad writing',\n",
       " 'dinosaurs',\n",
       " 'disaster',\n",
       " 'predictable',\n",
       " 'unrealistic',\n",
       " 'annoying kid',\n",
       " 'slasher',\n",
       " 'detective',\n",
       " '1930s',\n",
       " '1935',\n",
       " 'church',\n",
       " 'cotton',\n",
       " 'cotton farm',\n",
       " 'debt',\n",
       " 'extramarital affair',\n",
       " 'farm',\n",
       " 'female protagonist',\n",
       " 'ku klux klan',\n",
       " 'loss of husband',\n",
       " 'period piece',\n",
       " 'poverty',\n",
       " 'rural setting',\n",
       " 'nudity',\n",
       " 'dance',\n",
       " 'feel-good',\n",
       " 'motivational',\n",
       " 'terrible-acting',\n",
       " 'ghost story',\n",
       " 'haunted house',\n",
       " 'legal thriller',\n",
       " 'conspiracy theory',\n",
       " 'ufo',\n",
       " 'female bonding',\n",
       " 'exorcism',\n",
       " 'bad acting',\n",
       " 'haunted-house',\n",
       " 'jumpscares',\n",
       " 'siblings',\n",
       " 'artist',\n",
       " 'con man',\n",
       " 'doppelgänger',\n",
       " 'evil doppelgänger',\n",
       " 'loss of child',\n",
       " 'stephen king',\n",
       " 'zombies',\n",
       " 'breakup',\n",
       " 'feminist',\n",
       " 'friends',\n",
       " 'girlfriends',\n",
       " 'heartbreak',\n",
       " 'lesbian',\n",
       " 'film industry',\n",
       " 'hollywood',\n",
       " 'movie industry',\n",
       " 'visual effect',\n",
       " 'cyber-intimidation',\n",
       " 'hackers',\n",
       " 'hacking',\n",
       " 'internet',\n",
       " 'domestic violence',\n",
       " 'fun',\n",
       " 'kids',\n",
       " 'schizophrenia',\n",
       " 'sleep walking',\n",
       " 'lucid dreaming',\n",
       " 'banksy',\n",
       " 'hand-held camera',\n",
       " 'philosophy of art',\n",
       " 'street art',\n",
       " 'made me laugh',\n",
       " 'parody: horror film episodes',\n",
       " 'satire: plastic surgery',\n",
       " 'word play',\n",
       " 'sport business',\n",
       " 'childish',\n",
       " 'sexist',\n",
       " 'imprisonment',\n",
       " 'multi-dimensional characters',\n",
       " 'taking justice into own hands',\n",
       " 'illogical',\n",
       " 'nature',\n",
       " 'strong female character',\n",
       " 'comics',\n",
       " 'jack the ripper',\n",
       " 'johnny depp',\n",
       " 'serial killer',\n",
       " 'kyra sedgwick',\n",
       " 'french music',\n",
       " 'implausible',\n",
       " 'rachel mcadams',\n",
       " 'not spotting the obvious',\n",
       " 'realistic action',\n",
       " 'wonderful director',\n",
       " 'bing crosby',\n",
       " 'frank sinatra',\n",
       " 'louis armstrong',\n",
       " 'remake',\n",
       " 'evil-mastermind',\n",
       " 'fourth-wall',\n",
       " 'light',\n",
       " 'spy',\n",
       " 'clash of cultures',\n",
       " 'film within a film',\n",
       " 'life & death',\n",
       " 'little dialogue',\n",
       " 'mona achache',\n",
       " 'rebirth',\n",
       " 'deception',\n",
       " 'illusions',\n",
       " 'harrison ford',\n",
       " 'true to the book',\n",
       " 'great visuals',\n",
       " 'marvel cinematic universe',\n",
       " 'boy becomes a man',\n",
       " 'redemption',\n",
       " 'life philosophy',\n",
       " 'mozi philosophy',\n",
       " 'personal growth',\n",
       " 'strong romantic subplot',\n",
       " 'war strategy',\n",
       " 'adult',\n",
       " 'antihero',\n",
       " 'strong language',\n",
       " 'punk',\n",
       " 'art film',\n",
       " 'entertainment industry',\n",
       " 'k-pop',\n",
       " 'ending kinda ruined it',\n",
       " 'great performances',\n",
       " 'sucks',\n",
       " 'nice',\n",
       " 'jason lee',\n",
       " 'kevin smith',\n",
       " 'view askew',\n",
       " 'gene wilder',\n",
       " 'nick frost',\n",
       " 'simon pegg',\n",
       " 'games',\n",
       " 'bd-r',\n",
       " 'space action',\n",
       " 'hallucinatory',\n",
       " 'middle school',\n",
       " 'pixar',\n",
       " 'lady gaga',\n",
       " 'intense',\n",
       " 'courage',\n",
       " 'inspirational',\n",
       " 'reality',\n",
       " 'willpower',\n",
       " 'different',\n",
       " 'persistence',\n",
       " 'post apocalyptic',\n",
       " 'bradley cooper',\n",
       " 'concept',\n",
       " 'entertaining',\n",
       " 'human potential',\n",
       " 'screenplay',\n",
       " 'smart drugs',\n",
       " 'talent',\n",
       " 'trying hard',\n",
       " 'mathematicians',\n",
       " 'eddie redmayne',\n",
       " 'excellent acting',\n",
       " 'excellent screenplay',\n",
       " 'true love',\n",
       " 'revenge as motive',\n",
       " 'hilarious comedy',\n",
       " 'investigation',\n",
       " 'central intelligence agency',\n",
       " 'dvd',\n",
       " 'art house',\n",
       " 'love this movie',\n",
       " 'scaring greatsoundtrack davidlynch angelobadalamenti thriller',\n",
       " 'dialogs',\n",
       " 'nothing',\n",
       " 'short stories',\n",
       " 'mother-son relationship',\n",
       " 'cheesy',\n",
       " 'cheesiness',\n",
       " 'ridiculous',\n",
       " 'surprise ending',\n",
       " 'insanity',\n",
       " 'martin scorsese',\n",
       " 'psychological thriller',\n",
       " 'mind bending',\n",
       " 'spirit',\n",
       " 'freedom',\n",
       " 'hope',\n",
       " 'aardman',\n",
       " 'tear jerker',\n",
       " 'strong female lead',\n",
       " 'anime. miyazaki',\n",
       " 'tim burton',\n",
       " 'jack black',\n",
       " 'comic book adaption',\n",
       " 'marvel',\n",
       " 'masterpiece',\n",
       " 'stand up',\n",
       " 'vistavision',\n",
       " 'bruce willis',\n",
       " 'the twists in the plot',\n",
       " 'great ending',\n",
       " 'supernatural',\n",
       " 'morality',\n",
       " 'epic',\n",
       " 'historical',\n",
       " 'medieval',\n",
       " 'space opera',\n",
       " 'crime',\n",
       " 'american history',\n",
       " 'vietnam war',\n",
       " 'biblical',\n",
       " 'british',\n",
       " 'british comedy',\n",
       " 'controversial',\n",
       " 'monty python',\n",
       " 'akira kurosawa',\n",
       " 'french',\n",
       " 'stylish',\n",
       " 'farce',\n",
       " 'mockumentary',\n",
       " 'no dialogue',\n",
       " 'no narrative',\n",
       " 'edward d. wood jr.',\n",
       " \"so bad it's funny\",\n",
       " 'long',\n",
       " 'samurai',\n",
       " 'toshiro mifune',\n",
       " 'existentialism',\n",
       " 'scifi',\n",
       " 'survival horror',\n",
       " 'free speech',\n",
       " 'memory',\n",
       " 'anti-war',\n",
       " 'wilderness',\n",
       " 'deadpan',\n",
       " 'hip hop',\n",
       " 'lone hero',\n",
       " 'clever',\n",
       " 'penn jilette',\n",
       " 'sad ending',\n",
       " 'teller',\n",
       " 'comic book',\n",
       " 'author:jane austen',\n",
       " 'comedy of manners',\n",
       " 'norwegian',\n",
       " 'saami',\n",
       " 'skiing',\n",
       " 'brutal',\n",
       " 'vengeance',\n",
       " 'excellent use of dialogue',\n",
       " 'intellectual',\n",
       " 'ambiguous ending',\n",
       " 'depression',\n",
       " 'finnish',\n",
       " 'mythology',\n",
       " 'anti-authoritarian',\n",
       " 'edward snowden',\n",
       " 'informative',\n",
       " 'nonfiction',\n",
       " 'cute boys',\n",
       " 'cyborgs',\n",
       " 'nuclear war',\n",
       " 'high fantasy',\n",
       " 'wizards',\n",
       " 'scarlett johansson',\n",
       " 'fast paced action',\n",
       " '007',\n",
       " 'bond',\n",
       " 'james bond',\n",
       " 'sean connery',\n",
       " 'spy thriller',\n",
       " 'business',\n",
       " 'corporate america',\n",
       " 'michael douglas',\n",
       " 'stock market',\n",
       " 'us history',\n",
       " 'daniel craig',\n",
       " 'poker',\n",
       " 'spies',\n",
       " '1920s',\n",
       " 'f. scott fitzgerald',\n",
       " 'leonardo dicaprio',\n",
       " 'prostitution',\n",
       " 'fairy tale',\n",
       " 'bureaucracy',\n",
       " 'cerebral',\n",
       " 'satirical',\n",
       " 'neo-noir',\n",
       " 'astronomy',\n",
       " 'catastrophe',\n",
       " 'natural disaster',\n",
       " 'detention',\n",
       " 'cyberpunk',\n",
       " 'virtual reality',\n",
       " 'amazing cinematography',\n",
       " 'fascism',\n",
       " 'totalitarianism',\n",
       " 'h.p. lovecraft',\n",
       " 'lovecraftian mythology',\n",
       " 'philip k. dick',\n",
       " 'teleportation',\n",
       " 'alternate history',\n",
       " 'author:alan moore',\n",
       " 'danny mcbride',\n",
       " 'james franco',\n",
       " 'robert downey jr.',\n",
       " 'satire:hollywood insiders',\n",
       " 'atmosphere',\n",
       " 'steampunk',\n",
       " 'stylistic',\n",
       " 'good concept',\n",
       " 'saramago',\n",
       " 'artistic',\n",
       " 'escapism',\n",
       " 'hanging',\n",
       " 'victorian era',\n",
       " 'puzzle',\n",
       " 'sci fi',\n",
       " 'twists & turns',\n",
       " 'ending twist',\n",
       " 'mental institution',\n",
       " 'reality or imagination?',\n",
       " 'twisted ending',\n",
       " 'great humor',\n",
       " 'great screenplay',\n",
       " 'joss whedon',\n",
       " 'airships',\n",
       " 'anne hathaway',\n",
       " 'christian bale',\n",
       " 'atheism',\n",
       " 'david fincher',\n",
       " 'chaotic',\n",
       " 'meta',\n",
       " 'awkward situations',\n",
       " 'comedic horror',\n",
       " 'cured by love/hope',\n",
       " 'unconventional zombies',\n",
       " 'samuel l. jackson',\n",
       " 'entertaining concept',\n",
       " 'michael cera',\n",
       " 'gwyneth paltrow',\n",
       " 'humour',\n",
       " 'iron man',\n",
       " 'shane black',\n",
       " 'found footage',\n",
       " 'hard science fiction',\n",
       " 'lovecraftian',\n",
       " 'non-sequential',\n",
       " 'realistic sci-fi',\n",
       " 'visually stunning',\n",
       " 'dumb science',\n",
       " 'transhumanist',\n",
       " 'convoluted',\n",
       " 'cybernetics',\n",
       " 'low-budget sci fi',\n",
       " 'multiverse',\n",
       " 'parallel universe',\n",
       " 'dystopian',\n",
       " 'post apocalypse',\n",
       " 'tom hardy',\n",
       " 'exploited',\n",
       " 'fan service',\n",
       " 'plot recycling',\n",
       " 'recycled plot',\n",
       " 'x-men',\n",
       " 'afrofuturism',\n",
       " 'colonialism',\n",
       " 'mcu',\n",
       " 'busy plot',\n",
       " 'cliffhanger',\n",
       " 'genocide',\n",
       " 'slow burn',\n",
       " 'thanos',\n",
       " 'pop culture',\n",
       " 'steven spielberg',\n",
       " 'steven spilberg',\n",
       " 'cult',\n",
       " 'satanism',\n",
       " 'beautiful visuals',\n",
       " 'cerebral sci-fi',\n",
       " 'communication',\n",
       " 'extraterrestrial',\n",
       " 'fatalist',\n",
       " 'good cinematography',\n",
       " 'great aliens',\n",
       " 'language',\n",
       " 'linguistics',\n",
       " 'nonlinear timeline',\n",
       " 'scientist',\n",
       " 'sense of awe',\n",
       " 'thoughtful sci-fi',\n",
       " 'claustrophobic',\n",
       " 'body horror',\n",
       " 'mad doctor',\n",
       " 'power rangers',\n",
       " 'product placement',\n",
       " ...]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_tags_mod = data_tags.copy()\n",
    "\n",
    "# Deixando todas as tags minusculas para ter um formato mais uniforme\n",
    "data_tags_mod[\"tag\"] = data_tags_mod[\"tag\"].str.lower()\n",
    "\n",
    "# Separando todas as tags em uma lista\n",
    "tags_list = data_tags_mod.tag.unique()\n",
    "tags_list = tags_list.tolist()\n",
    "tags_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:43:44.043982Z",
     "start_time": "2021-06-24T15:43:44.033980Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "65465"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(tags_list)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### TRABALHANDO COM LINGUAGEM NATURAL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:44:03.080927Z",
     "start_time": "2021-06-24T15:44:01.850357Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>tags</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>owned imdb top 250 pixar pixar time travel children comedy funny witty rated-g animation pixar computer animation good cartoon chindren pixar friendship bright daring rescues fanciful heroic mission humorous light rousing toys come to life unlikely friendships warm witty animation disney friendship pixar boy boy next door bullying friends friendship jealousy martial arts mission neighborhood new toy pixar rescue resourcefulness rivalry toy toy comes to life walkie talkie clever animation cle...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>robin williams time travel fantasy based on children's book board game disappearance giant insect new home recluse bad cgi kid flick scary time travel animals bad cgi board game not for kids robin williams time travel fantasy magic board game robin williams kid flick adventure animals board game family fantasy fiction robin williams time travel fantasy robin williams adapted from:book childish children kid flick time travel animals bad cgi based on a book board game chris van allsburg animal...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>funny best friend duringcreditsstinger fishing old man sequel fever moldy old sequel no_fa_ganes jack lemmon walter matthau grun running comedinha de velhinhos engraãƒâ§ada old people that is actually funny howard deutch clv sequel comedinha de velhinhos engraã§ada moldy old ann margaret burgess meredith daryl hannah fishing good soundtrack jack lemmon sophia loren walter matthau</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>based on novel or book chick flick divorce interracial relationship single mother clv chick flick revenge characters characters chick flick</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>aging baby confidence contraception daughter gynecologist midlife crisis parent child relationship pregnancy sequel fever fantasy steve martin comedy touching clv steve martin family pregnancy wedding diane keaton family sequel steve martin wedding</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId  \\\n",
       "0        1   \n",
       "1        2   \n",
       "2        3   \n",
       "3        4   \n",
       "4        5   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                  tags  \n",
       "0  owned imdb top 250 pixar pixar time travel children comedy funny witty rated-g animation pixar computer animation good cartoon chindren pixar friendship bright daring rescues fanciful heroic mission humorous light rousing toys come to life unlikely friendships warm witty animation disney friendship pixar boy boy next door bullying friends friendship jealousy martial arts mission neighborhood new toy pixar rescue resourcefulness rivalry toy toy comes to life walkie talkie clever animation cle...  \n",
       "1  robin williams time travel fantasy based on children's book board game disappearance giant insect new home recluse bad cgi kid flick scary time travel animals bad cgi board game not for kids robin williams time travel fantasy magic board game robin williams kid flick adventure animals board game family fantasy fiction robin williams time travel fantasy robin williams adapted from:book childish children kid flick time travel animals bad cgi based on a book board game chris van allsburg animal...  \n",
       "2                                                                                                                       funny best friend duringcreditsstinger fishing old man sequel fever moldy old sequel no_fa_ganes jack lemmon walter matthau grun running comedinha de velhinhos engraãƒâ§ada old people that is actually funny howard deutch clv sequel comedinha de velhinhos engraã§ada moldy old ann margaret burgess meredith daryl hannah fishing good soundtrack jack lemmon sophia loren walter matthau  \n",
       "3                                                                                                                                                                                                                                                                                                                                                                          based on novel or book chick flick divorce interracial relationship single mother clv chick flick revenge characters characters chick flick  \n",
       "4                                                                                                                                                                                                                                                             aging baby confidence contraception daughter gynecologist midlife crisis parent child relationship pregnancy sequel fever fantasy steve martin comedy touching clv steve martin family pregnancy wedding diane keaton family sequel steve martin wedding  "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Primeiro juntando todas as tags em uma unica linha\n",
    "data_tags_grouped = data_tags_mod.copy()\n",
    "\n",
    "data_tags_grouped.fillna(\"\", inplace=True)\n",
    "\n",
    "# Agrupando as tags em uma linha só\n",
    "data_tags_grouped = pd.DataFrame(data_tags_grouped.groupby('movieId')['tag'].apply(lambda x: \"%s\" %' '.join(x)))\n",
    "\n",
    "# Resetando index\n",
    "data_tags_grouped.reset_index(inplace=True)\n",
    "data_tags_grouped.rename(columns={\"tag\":\"tags\"}, inplace=True)\n",
    "pd.set_option('display.max_colwidth', 500)\n",
    "data_tags_grouped.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### PROCESSANDO OS METADADOS FORMADOS PELA JUNÇÃO DE TAGS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:44:07.993349Z",
     "start_time": "2021-06-24T15:44:06.913656Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package punkt to\n",
      "[nltk_data]     C:\\Users\\SeaSkyThe\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package punkt is already up-to-date!\n",
      "[nltk_data] Downloading package averaged_perceptron_tagger to\n",
      "[nltk_data]     C:\\Users\\SeaSkyThe\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package averaged_perceptron_tagger is already up-to-\n",
      "[nltk_data]       date!\n",
      "[nltk_data] Downloading package wordnet to\n",
      "[nltk_data]     C:\\Users\\SeaSkyThe\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package wordnet is already up-to-date!\n",
      "[nltk_data] Downloading package stopwords to\n",
      "[nltk_data]     C:\\Users\\SeaSkyThe\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    }
   ],
   "source": [
    "pd.set_option('display.max_colwidth', 50)\n",
    "\n",
    "# SETANDO CONFIGURACOES PARA PROCESSAR AS TAGS\n",
    "import nltk\n",
    "nltk.download('punkt')\n",
    "nltk.download('averaged_perceptron_tagger')\n",
    "nltk.download('wordnet')\n",
    "  \n",
    "from nltk.stem import WordNetLemmatizer\n",
    "lemmatizer = WordNetLemmatizer()\n",
    "  \n",
    "from nltk.corpus import stopwords\n",
    "nltk.download('stopwords')\n",
    "stop_words = set(stopwords.words('english'))\n",
    "  \n",
    "VERB_CODES = {'VB', 'VBD', 'VBG', 'VBN', 'VBP', 'VBZ'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:34.122689Z",
     "start_time": "2021-06-24T15:44:08.698339Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>tags</th>\n",
       "      <th>tags_processed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>45246</th>\n",
       "      <td>208813</td>\n",
       "      <td>might like</td>\n",
       "      <td>might like</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45247</th>\n",
       "      <td>208933</td>\n",
       "      <td>black and white deal with the devil</td>\n",
       "      <td>black white deal devil</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45248</th>\n",
       "      <td>209035</td>\n",
       "      <td>computer animation japan mass behavior mass sc...</td>\n",
       "      <td>computer animation japan mass behavior mass sc...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45249</th>\n",
       "      <td>209037</td>\n",
       "      <td>chameleon computer animation gluttony humorous...</td>\n",
       "      <td>chameleon computer animation gluttony humorous...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45250</th>\n",
       "      <td>209063</td>\n",
       "      <td>black education friends schools independent sc...</td>\n",
       "      <td>black education friend school independent scho...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       movieId                                               tags  \\\n",
       "45246   208813                                         might like   \n",
       "45247   208933                black and white deal with the devil   \n",
       "45248   209035  computer animation japan mass behavior mass sc...   \n",
       "45249   209037  chameleon computer animation gluttony humorous...   \n",
       "45250   209063  black education friends schools independent sc...   \n",
       "\n",
       "                                          tags_processed  \n",
       "45246                                         might like  \n",
       "45247                             black white deal devil  \n",
       "45248  computer animation japan mass behavior mass sc...  \n",
       "45249  chameleon computer animation gluttony humorous...  \n",
       "45250  black education friend school independent scho...  "
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# CRIANDO FUNCAO PARA PROCESSAR AS TAGS, TIRANDO CONTRAÇÕES DE PALAVRAS (n't = not)\n",
    "\n",
    "def preprocess_sentences(text):\n",
    "    # garantindo que o texto seja minusculo (apesar de já termos feito isso)\n",
    "    text = text.lower()\n",
    "    temp_sent = []\n",
    "    words = nltk.word_tokenize(text)\n",
    "    tags = nltk.pos_tag(words)\n",
    "    for i, word in enumerate(words):\n",
    "        if tags[i][1] in VERB_CODES:\n",
    "            lemmatized = lemmatizer.lemmatize(word, 'v')\n",
    "        else:\n",
    "            lemmatized = lemmatizer.lemmatize(word)\n",
    "        if lemmatized not in stop_words and lemmatized.isalpha():\n",
    "            temp_sent.append(lemmatized)\n",
    "    \n",
    "    finalsent = ' '.join(temp_sent)\n",
    "    finalsent = finalsent.replace(\"n't\", \" not\")\n",
    "    finalsent = finalsent.replace(\"'m\", \" am\")\n",
    "    finalsent = finalsent.replace(\"'s\", \" is\")\n",
    "    finalsent = finalsent.replace(\"'re\", \" are\")\n",
    "    finalsent = finalsent.replace(\"'ll\", \" will\")\n",
    "    finalsent = finalsent.replace(\"'ve\", \" have\")\n",
    "    finalsent = finalsent.replace(\"'d\", \" would\")\n",
    "    return finalsent\n",
    "\n",
    "data_tags_grouped[\"tags_processed\"]= data_tags_grouped[\"tags\"].apply(preprocess_sentences)\n",
    "data_tags_grouped.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:34.138692Z",
     "start_time": "2021-06-24T15:46:34.123690Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>tags</th>\n",
       "      <th>tags_processed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>owned imdb top 250 pixar pixar time travel children comedy funny witty rated-g animation pixar computer animation good cartoon chindren pixar friendship bright daring rescues fanciful heroic mission humorous light rousing toys come to life unlikely friendships warm witty animation disney friendship pixar boy boy next door bullying friends friendship jealousy martial arts mission neighborhood new toy pixar rescue resourcefulness rivalry toy toy comes to life walkie talkie clever animation cle...</td>\n",
       "      <td>imdb top pixar pixar time travel child comedy funny witty animation pixar computer animation good cartoon chindren pixar friendship bright dare rescue fanciful heroic mission humorous light rouse toy come life unlikely friendship warm witty animation disney friendship pixar boy boy next door bully friend friendship jealousy martial art mission neighborhood new toy pixar rescue resourcefulness rivalry toy toy come life walkie talkie clever animation clever friendship funny tom hank witty pixa...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>robin williams time travel fantasy based on children's book board game disappearance giant insect new home recluse bad cgi kid flick scary time travel animals bad cgi board game not for kids robin williams time travel fantasy magic board game robin williams kid flick adventure animals board game family fantasy fiction robin williams time travel fantasy robin williams adapted from:book childish children kid flick time travel animals bad cgi based on a book board game chris van allsburg animal...</td>\n",
       "      <td>robin williams time travel fantasy base child book board game disappearance giant insect new home recluse bad cgi kid flick scary time travel animal bad cgi board game kid robin williams time travel fantasy magic board game robin williams kid flick adventure animal board game family fantasy fiction robin williams time travel fantasy robin williams adapt book childish child kid flick time travel animal bad cgi base book board game chris van allsburg animal fantasy robin williams scary robin w...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>funny best friend duringcreditsstinger fishing old man sequel fever moldy old sequel no_fa_ganes jack lemmon walter matthau grun running comedinha de velhinhos engraãƒâ§ada old people that is actually funny howard deutch clv sequel comedinha de velhinhos engraã§ada moldy old ann margaret burgess meredith daryl hannah fishing good soundtrack jack lemmon sophia loren walter matthau</td>\n",
       "      <td>funny best friend duringcreditsstinger fish old man sequel fever moldy old sequel jack lemmon walter matthau grun run comedinha de velhinhos old people actually funny howard deutch clv sequel comedinha de velhinhos moldy old ann margaret burgess meredith daryl hannah fish good soundtrack jack lemmon sophia loren walter matthau</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>based on novel or book chick flick divorce interracial relationship single mother clv chick flick revenge characters characters chick flick</td>\n",
       "      <td>base novel book chick flick divorce interracial relationship single mother clv chick flick revenge character character chick flick</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>aging baby confidence contraception daughter gynecologist midlife crisis parent child relationship pregnancy sequel fever fantasy steve martin comedy touching clv steve martin family pregnancy wedding diane keaton family sequel steve martin wedding</td>\n",
       "      <td>age baby confidence contraception daughter gynecologist midlife crisis parent child relationship pregnancy sequel fever fantasy steve martin comedy touch clv steve martin family pregnancy wed diane keaton family sequel steve martin wedding</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>imdb top 250 great acting realistic action suspense al pacino atmospheric bank robbery crime great acting realistic action robert de niro suspense tense val kilmer bank job bank robbery crime dialogue heist heist movie long al pacino atmospheric realistic action robert de niro bank robbery los angeles realistic action al pacino atmospheric michael mann realistic action robert de niro visceral atmospheric bleak slick somber tense visceral witty bank job betrayal cat and mouse chase crime epic...</td>\n",
       "      <td>imdb top great act realistic action suspense al pacino atmospheric bank robbery crime great act realistic action robert de niro suspense tense val kilmer bank job bank robbery crime dialogue heist heist movie long al pacino atmospheric realistic action robert de niro bank robbery los angeles realistic action al pacino atmospheric michael mann realistic action robert de niro visceral atmospheric bleak slick somber tense visceral witty bank job betrayal cat mouse chase crime epic criminal mast...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>remake chauffeur fusion long island millionaire paris sibling relationship based on a play harrison ford paris romance sibling relationship based on a play ugly duckling relationships great cast harrison ford romance love story romance no_fa_ganes remake drama finding yourself great cast greg kinnear julia ormond love story remadeas:sabrina(1995) remakeof:sabrina(1954) upstairs-downstairs as good maybe better than original remake clv romance as good maybe better than original harrison ford g...</td>\n",
       "      <td>remake chauffeur fusion long island millionaire paris sibling relationship base play harrison ford paris romance sibling relationship base play ugly duckling relationship great cast harrison ford romance love story romance remake drama find great cast greg kinnear julia ormond love story remadeas sabrina remakeof sabrina good maybe better original remake clv romance good maybe better original harrison ford greg kinnear romance harrison ford greg kinnear remake romance chick flick remake harr...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>based on a book mark twain peter hewitt adapted from:book author:mark twain library system disney</td>\n",
       "      <td>base book mark twain peter hewitt adapt book author mark twain library system disney</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>explosive hostage terrorist vice president jean-claude van damme jean-claude van damme peter hyams clv action jean-claude van damme peter hyams can't remember jean-claude van damme</td>\n",
       "      <td>explosive hostage terrorist vice president van damme van damme peter hyams clv action van damme peter hyams ca remember van damme</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>007 bond boys with toys gadgets secret service sequel spies violence computer virus cossack cuba electromagnetic pulse falsely accused kgb red army satellite secret base secret identity secret intelligence service special car st. petersburg russia time bomb 007 james bond 007 bond 007 action james bond pierce brosnan pseudo rated i wanted the bad guy to win.  sean bean out-bonds bond. best of the brosnon bonds tank chase scene caribbean espionage franchise james bond latin america puerto ric...</td>\n",
       "      <td>bond boy toy gadget secret service sequel spy violence computer virus cossack cuba electromagnetic pulse falsely accuse kgb red army satellite secret base secret identity secret intelligence service special car petersburg russia time bomb james bond bond action james bond pierce brosnan pseudo rat want bad guy win sean bean bond best brosnon bond tank chase scene caribbean espionage franchise jam bond latin america puerto rico bond espionage jam bond bond pierce brosnan bond bond bond action...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>11</td>\n",
       "      <td>romance white house new love usa president white house widower wildlife conservation aaron sorkin romance politics single parents us president aaron sorkin aaron sorkin politics romance no_fa_ganes amazing kung fu decent flick annette bening ethics great cast great performances great story michael douglas morals rob reiner clv drama romance president girlie movie romance aaron sorkin michael douglas politics morals politics president cute dvd collection president us president white house aar...</td>\n",
       "      <td>romance white house new love usa president white house widower wildlife conservation aaron sorkin romance politics single parent u president aaron sorkin aaron sorkin politics romance amaze kung fu decent flick annette bening ethic great cast great performance great story michael douglas moral rob reiner clv drama romance president girlie movie romance aaron sorkin michael douglas politics moral politics president cute dvd collection president u president white house aaron sorkin politics bi...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>12</td>\n",
       "      <td>dracula spoof mel brooks gothic leslie nielsen funny spoof leslie nielsen mel brooks leslie nielsen bd-r leslie nielsen leslie nielsen mel brooks spoof parody farce spoof vampires spoof vampire</td>\n",
       "      <td>dracula spoof mel brook gothic leslie nielsen funny spoof leslie nielsen mel brook leslie nielsen leslie nielsen leslie nielsen mel brook spoof parody farce spoof vampire spoof vampire</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>13</td>\n",
       "      <td>ei muista alaska bear attack dog dog sled dog-sledding race frozen lake goose pet wolf james horner dogs dogsled sort of boring animated dogs sled dogs talking animals simon wells itaege wolves animation dogs talking animals animated dogs sled dogs talking animals</td>\n",
       "      <td>ei muista alaska bear attack dog dog sled race frozen lake goose pet wolf james horner dog dogsled sort bore animated dog sled dog talk animal simon well itaege wolf animation dog talk animal animate dog sled dog talk animal</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>14</td>\n",
       "      <td>biography government historical figure presidential election usa president watergate scandal corruption politics true story us president politics oliver stone biographical biopic character based on real person:richard nixon us president cinematographer:robert richardson clv watergate president political drama based on a true story biographical oliver stone politics historical opinion tumey's vhs own oliver stone anthony hopkins drama based on a true story long</td>\n",
       "      <td>biography government historical figure presidential election usa president watergate scandal corruption politics true story u president politics oliver stone biographical biopic character base real person richard nixon u president cinematographer robert richardson clv watergate president political drama base true story biographical oliver stone politics historical opinion tumey vhs oliver stone anthony hopkins drama base true story long</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>15</td>\n",
       "      <td>exotic island map pirate scalp ship treasure big budget treasure pirates action adventure better than i remembered frank langella geena davis humor pirate pirates sword fight treasure treasure hunt treasure pirates pirates clv treasure pirates sexuality treasure treasure pirates pirate treasure treasure bad acting geena davis pirates treasure pirate pirates</td>\n",
       "      <td>exotic island map pirate scalp ship treasure big budget treasure pirate action adventure better remember frank langella geena davis humor pirate pirate sword fight treasure treasure hunt treasure pirate pirate clv treasure pirate sexuality treasure treasure pirate pirate treasure treasure bad act geena davis pirate treasure pirate pirate</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    movieId  \\\n",
       "0         1   \n",
       "1         2   \n",
       "2         3   \n",
       "3         4   \n",
       "4         5   \n",
       "5         6   \n",
       "6         7   \n",
       "7         8   \n",
       "8         9   \n",
       "9        10   \n",
       "10       11   \n",
       "11       12   \n",
       "12       13   \n",
       "13       14   \n",
       "14       15   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   tags  \\\n",
       "0   owned imdb top 250 pixar pixar time travel children comedy funny witty rated-g animation pixar computer animation good cartoon chindren pixar friendship bright daring rescues fanciful heroic mission humorous light rousing toys come to life unlikely friendships warm witty animation disney friendship pixar boy boy next door bullying friends friendship jealousy martial arts mission neighborhood new toy pixar rescue resourcefulness rivalry toy toy comes to life walkie talkie clever animation cle...   \n",
       "1   robin williams time travel fantasy based on children's book board game disappearance giant insect new home recluse bad cgi kid flick scary time travel animals bad cgi board game not for kids robin williams time travel fantasy magic board game robin williams kid flick adventure animals board game family fantasy fiction robin williams time travel fantasy robin williams adapted from:book childish children kid flick time travel animals bad cgi based on a book board game chris van allsburg animal...   \n",
       "2                                                                                                                        funny best friend duringcreditsstinger fishing old man sequel fever moldy old sequel no_fa_ganes jack lemmon walter matthau grun running comedinha de velhinhos engraãƒâ§ada old people that is actually funny howard deutch clv sequel comedinha de velhinhos engraã§ada moldy old ann margaret burgess meredith daryl hannah fishing good soundtrack jack lemmon sophia loren walter matthau   \n",
       "3                                                                                                                                                                                                                                                                                                                                                                           based on novel or book chick flick divorce interracial relationship single mother clv chick flick revenge characters characters chick flick   \n",
       "4                                                                                                                                                                                                                                                              aging baby confidence contraception daughter gynecologist midlife crisis parent child relationship pregnancy sequel fever fantasy steve martin comedy touching clv steve martin family pregnancy wedding diane keaton family sequel steve martin wedding   \n",
       "5   imdb top 250 great acting realistic action suspense al pacino atmospheric bank robbery crime great acting realistic action robert de niro suspense tense val kilmer bank job bank robbery crime dialogue heist heist movie long al pacino atmospheric realistic action robert de niro bank robbery los angeles realistic action al pacino atmospheric michael mann realistic action robert de niro visceral atmospheric bleak slick somber tense visceral witty bank job betrayal cat and mouse chase crime epic...   \n",
       "6   remake chauffeur fusion long island millionaire paris sibling relationship based on a play harrison ford paris romance sibling relationship based on a play ugly duckling relationships great cast harrison ford romance love story romance no_fa_ganes remake drama finding yourself great cast greg kinnear julia ormond love story remadeas:sabrina(1995) remakeof:sabrina(1954) upstairs-downstairs as good maybe better than original remake clv romance as good maybe better than original harrison ford g...   \n",
       "7                                                                                                                                                                                                                                                                                                                                                                                                                     based on a book mark twain peter hewitt adapted from:book author:mark twain library system disney   \n",
       "8                                                                                                                                                                                                                                                                                                                                  explosive hostage terrorist vice president jean-claude van damme jean-claude van damme peter hyams clv action jean-claude van damme peter hyams can't remember jean-claude van damme   \n",
       "9   007 bond boys with toys gadgets secret service sequel spies violence computer virus cossack cuba electromagnetic pulse falsely accused kgb red army satellite secret base secret identity secret intelligence service special car st. petersburg russia time bomb 007 james bond 007 bond 007 action james bond pierce brosnan pseudo rated i wanted the bad guy to win.  sean bean out-bonds bond. best of the brosnon bonds tank chase scene caribbean espionage franchise james bond latin america puerto ric...   \n",
       "10  romance white house new love usa president white house widower wildlife conservation aaron sorkin romance politics single parents us president aaron sorkin aaron sorkin politics romance no_fa_ganes amazing kung fu decent flick annette bening ethics great cast great performances great story michael douglas morals rob reiner clv drama romance president girlie movie romance aaron sorkin michael douglas politics morals politics president cute dvd collection president us president white house aar...   \n",
       "11                                                                                                                                                                                                                                                                                                                    dracula spoof mel brooks gothic leslie nielsen funny spoof leslie nielsen mel brooks leslie nielsen bd-r leslie nielsen leslie nielsen mel brooks spoof parody farce spoof vampires spoof vampire   \n",
       "12                                                                                                                                                                                                                                             ei muista alaska bear attack dog dog sled dog-sledding race frozen lake goose pet wolf james horner dogs dogsled sort of boring animated dogs sled dogs talking animals simon wells itaege wolves animation dogs talking animals animated dogs sled dogs talking animals   \n",
       "13                                     biography government historical figure presidential election usa president watergate scandal corruption politics true story us president politics oliver stone biographical biopic character based on real person:richard nixon us president cinematographer:robert richardson clv watergate president political drama based on a true story biographical oliver stone politics historical opinion tumey's vhs own oliver stone anthony hopkins drama based on a true story long   \n",
       "14                                                                                                                                              exotic island map pirate scalp ship treasure big budget treasure pirates action adventure better than i remembered frank langella geena davis humor pirate pirates sword fight treasure treasure hunt treasure pirates pirates clv treasure pirates sexuality treasure treasure pirates pirate treasure treasure bad acting geena davis pirates treasure pirate pirates   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                         tags_processed  \n",
       "0   imdb top pixar pixar time travel child comedy funny witty animation pixar computer animation good cartoon chindren pixar friendship bright dare rescue fanciful heroic mission humorous light rouse toy come life unlikely friendship warm witty animation disney friendship pixar boy boy next door bully friend friendship jealousy martial art mission neighborhood new toy pixar rescue resourcefulness rivalry toy toy come life walkie talkie clever animation clever friendship funny tom hank witty pixa...  \n",
       "1   robin williams time travel fantasy base child book board game disappearance giant insect new home recluse bad cgi kid flick scary time travel animal bad cgi board game kid robin williams time travel fantasy magic board game robin williams kid flick adventure animal board game family fantasy fiction robin williams time travel fantasy robin williams adapt book childish child kid flick time travel animal bad cgi base book board game chris van allsburg animal fantasy robin williams scary robin w...  \n",
       "2                                                                                                                                                                              funny best friend duringcreditsstinger fish old man sequel fever moldy old sequel jack lemmon walter matthau grun run comedinha de velhinhos old people actually funny howard deutch clv sequel comedinha de velhinhos moldy old ann margaret burgess meredith daryl hannah fish good soundtrack jack lemmon sophia loren walter matthau  \n",
       "3                                                                                                                                                                                                                                                                                                                                                                                    base novel book chick flick divorce interracial relationship single mother clv chick flick revenge character character chick flick  \n",
       "4                                                                                                                                                                                                                                                                       age baby confidence contraception daughter gynecologist midlife crisis parent child relationship pregnancy sequel fever fantasy steve martin comedy touch clv steve martin family pregnancy wed diane keaton family sequel steve martin wedding  \n",
       "5   imdb top great act realistic action suspense al pacino atmospheric bank robbery crime great act realistic action robert de niro suspense tense val kilmer bank job bank robbery crime dialogue heist heist movie long al pacino atmospheric realistic action robert de niro bank robbery los angeles realistic action al pacino atmospheric michael mann realistic action robert de niro visceral atmospheric bleak slick somber tense visceral witty bank job betrayal cat mouse chase crime epic criminal mast...  \n",
       "6   remake chauffeur fusion long island millionaire paris sibling relationship base play harrison ford paris romance sibling relationship base play ugly duckling relationship great cast harrison ford romance love story romance remake drama find great cast greg kinnear julia ormond love story remadeas sabrina remakeof sabrina good maybe better original remake clv romance good maybe better original harrison ford greg kinnear romance harrison ford greg kinnear remake romance chick flick remake harr...  \n",
       "7                                                                                                                                                                                                                                                                                                                                                                                                                                  base book mark twain peter hewitt adapt book author mark twain library system disney  \n",
       "8                                                                                                                                                                                                                                                                                                                                                                                     explosive hostage terrorist vice president van damme van damme peter hyams clv action van damme peter hyams ca remember van damme  \n",
       "9   bond boy toy gadget secret service sequel spy violence computer virus cossack cuba electromagnetic pulse falsely accuse kgb red army satellite secret base secret identity secret intelligence service special car petersburg russia time bomb james bond bond action james bond pierce brosnan pseudo rat want bad guy win sean bean bond best brosnon bond tank chase scene caribbean espionage franchise jam bond latin america puerto rico bond espionage jam bond bond pierce brosnan bond bond bond action...  \n",
       "10  romance white house new love usa president white house widower wildlife conservation aaron sorkin romance politics single parent u president aaron sorkin aaron sorkin politics romance amaze kung fu decent flick annette bening ethic great cast great performance great story michael douglas moral rob reiner clv drama romance president girlie movie romance aaron sorkin michael douglas politics moral politics president cute dvd collection president u president white house aaron sorkin politics bi...  \n",
       "11                                                                                                                                                                                                                                                                                                                             dracula spoof mel brook gothic leslie nielsen funny spoof leslie nielsen mel brook leslie nielsen leslie nielsen leslie nielsen mel brook spoof parody farce spoof vampire spoof vampire  \n",
       "12                                                                                                                                                                                                                                                                                     ei muista alaska bear attack dog dog sled race frozen lake goose pet wolf james horner dog dogsled sort bore animated dog sled dog talk animal simon well itaege wolf animation dog talk animal animate dog sled dog talk animal  \n",
       "13                                                             biography government historical figure presidential election usa president watergate scandal corruption politics true story u president politics oliver stone biographical biopic character base real person richard nixon u president cinematographer robert richardson clv watergate president political drama base true story biographical oliver stone politics historical opinion tumey vhs oliver stone anthony hopkins drama base true story long  \n",
       "14                                                                                                                                                                  exotic island map pirate scalp ship treasure big budget treasure pirate action adventure better remember frank langella geena davis humor pirate pirate sword fight treasure treasure hunt treasure pirate pirate clv treasure pirate sexuality treasure treasure pirate pirate treasure treasure bad act geena davis pirate treasure pirate pirate  "
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.set_option('display.max_colwidth', 500)\n",
    "data_tags_grouped.head(15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:34.341692Z",
     "start_time": "2021-06-24T15:46:34.139693Z"
    }
   },
   "outputs": [],
   "source": [
    "# Salvando arquivo de tags processadas\n",
    "data_tags_grouped.to_pickle(PATH_TO_TAGS_PROCESSED_CB_FILE)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-23T22:37:05.659760Z",
     "start_time": "2021-06-23T22:37:05.620751Z"
    }
   },
   "source": [
    "#### O resto é feito no protótipo 2 (Vetorização e Criação do modelo)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Separando da lista de tags, apenas as que são relevantes, o critério será explicado posteriormente\n",
    "\n",
    "## OBS: isso foi feito acima de maneira automatica com tratamento de linguagem natural"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-23T20:47:23.013509Z",
     "start_time": "2021-06-23T20:47:14.193510Z"
    }
   },
   "outputs": [],
   "source": [
    "tags_list_mod = tags_list.copy()\n",
    "tags_list_mod = [str(i) for i in tags_list_mod]\n",
    "\n",
    "for tag in tags_list_mod:\n",
    "    # Removendo parenteses e seus conteudos\n",
    "    corrected_tag = re.sub(r' \\([^)]*\\)', '', tag)\n",
    "    #print(corrected_tag)\n",
    "    # \"based\" é um palavra comum de estar nas tags, e vai evitar cair em outros ifs\n",
    "    if 'based' in corrected_tag:\n",
    "        tags_list_mod[tags_list_mod.index(tag)] = corrected_tag\n",
    "        continue\n",
    "    # esse é para manter a tag sci-fi, e evita-la de ser removida no proximo if\n",
    "    if '-' in corrected_tag:\n",
    "        tags_list_mod[tags_list_mod.index(tag)] = corrected_tag\n",
    "        continue\n",
    "    \n",
    "   \n",
    "    if re.findall(r'\\b\\w{2}\\b', corrected_tag):\n",
    "        #todas as palavras de duas letras serão substituidas por NaN, que serão removidos depois\n",
    "        tags_list_mod[tags_list_mod.index(tag)] = 'nan'\n",
    "    \n",
    "    elif re.findall(r'\\b\\w{1}\\b', corrected_tag):\n",
    "        #todas as palavras de uma letra serão substituidas por NaN, que serão removidos depois\n",
    "        tags_list_mod[tags_list_mod.index(tag)] = 'nan'\n",
    "    \n",
    "    #remover as tags que são sites\n",
    "    elif(\"www\" in corrected_tag):\n",
    "        tags_list_mod[tags_list_mod.index(tag)] = 'nan'\n",
    "        \n",
    "    elif tag == corrected_tag:\n",
    "        #evitar substituição desnecessária\n",
    "        continue\n",
    "    \n",
    "    else:\n",
    "        tags_list_mod[tags_list_mod.index(tag)] = corrected_tag\n",
    "        pass\n",
    "\n",
    "tags_list_mod = [x for x in tags_list_mod if x != 'nan']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-23T20:55:53.301215Z",
     "start_time": "2021-06-23T20:55:53.192191Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "written by a 14-year-old boy\n",
      "stereoscopic 3-d\n",
      "55 movies every kid should see--entertainment weekly\n",
      "3-hours of sex and drugs\n",
      "pg-13\n",
      "pg-13 horror\n",
      "bb-8\n",
      "dimensionalized 2-d to 3-d\n",
      "rated-pg-13\n",
      "bosnian war of 1992-1995\n",
      "super-8\n",
      "based on 4-koma\n",
      "nc-17\n",
      "2-dimensional\n",
      "pg-13:sci-fi action violence\n",
      "qotd is the 3rd book - they squeezed the vampire lestat into 20 minutes of film\n",
      "notable nc-17\n",
      "watered-down pg-13 horror\n",
      "car: austin-healey 3000\n",
      "the first film to receive nc-17 rating\n",
      "pg-13:brief strong language\n",
      "pg-13:some sexuality\n",
      "pg-13:thematic elements\n",
      "r2-d2\n",
      "90s sci-fi\n",
      "pg-13:nudity\n",
      "attack of the 50-foot whatever\n",
      "the 5-second rule never applies near animals or in zoos\n",
      "karl urban needs to pick better projects & he needs 2 pick-up his sword\n",
      "super-35 blow-up\n",
      "28 days later rip-off\n",
      "blu-ray 3d\n",
      "70s sci-fi\n",
      "6.3-filmaffinity\n",
      "pg-13:some sexual material\n",
      "pg-13:sequences of sci-fi action and violence\n",
      "9-11\n",
      "c-3po\n",
      "7.5-filmaffinity\n",
      "7.2-filmaffinity\n",
      "6,5-filmaffinity\n",
      "8.7-filmaffinity\n",
      "7.4-filmaffinity\n",
      "8.4-filmaffinity\n",
      "8.2-filmaffinity\n",
      "8.3-filmaffinity\n",
      "7.6-filmaffinity\n",
      "6.5-filmaffinity\n",
      "7.0-filmaffinity\n",
      "6.8-filmaffinity\n",
      "7.1-filmaffinity\n",
      "7.8-filmaffinity\n",
      "7.3-filmaffinity\n",
      "6,2-filmaffinity\n",
      "5.7-filmaffinity\n",
      "7.7-filmaffinity\n",
      "6.7-filmaffinity\n",
      "5,4-filmaffinity\n",
      "6.4-filmaffinity\n",
      "6.2-filmaffinity\n",
      "4.5-filmafinity\n",
      "6.1-filmaffinity\n",
      "5.5-filmaffinity\n",
      "5.8-filmaffinity\n",
      "5.9-filmaffinity\n",
      "5.3-filmaffinity\n",
      "5.1-filmaffinity\n",
      "5-filmaffinity\n",
      "5,6-filmafinity\n",
      "5,2-filmafinity\n",
      "2001-like\n",
      "pg-13:brief sensuality\n",
      "2015-08\n",
      "06 oscar nominated best movie - foreign language\n",
      "universal 3-d\n",
      "50s sci-fi\n",
      "b-25 mitchell\n",
      "p-40 warhawk\n",
      "su-22 fitter\n",
      "007-like\n",
      "good use of 3-d\n",
      "50s-60s popcorn movie\n",
      "pre-20th century period\n",
      "pg-13:intense sequences of action and violence\n",
      "remake of 8-and-a-half\n",
      "\"singing by non-singers\" in the nyer 24/31 dec 2012\n",
      "see also:bubba ho-tep(2002)\n",
      "would have been 4.5 if not for the schmaltzy tough-guy-in-front-of-fireworks moment.\n",
      "11-5-2007\n",
      "2009 reissue in stereoscopic 3-d\n",
      "imax 3-d\n",
      "super 1.85 blow-up\n",
      "dimensionalized 2-d to 3-d for reissue in 2011\n",
      "dimensionalized 2-d to 3-d for reissue in 2013\n",
      "imax dmr 3-d reissue in 2013\n",
      "2006 reissue in stereoscopic 3-d\n",
      "dimension-150\n",
      "dimensionalized 2-d to 3-d for reissue in 2012\n",
      "imax dmr 3-d reissue in 2012\n",
      "anaglyphic 3-d sequences\n",
      "3-depix\n",
      "nwave 3-d\n",
      "iwerks 3-d\n",
      "mcnabb 3-d\n",
      "natural vision 3-dimension\n",
      "imax dmr 3-d\n",
      "columbia 3-d\n",
      "stereoscopic 3-d sequences\n",
      "bd-video 3d\n",
      "fusion 3-d\n",
      "tru 3-d\n",
      "anaglyphic 3-d\n",
      "is it 5/5-worthy?\n",
      "2 or 3 hilarious parts - but too boring to see again\n",
      "2020-2 relook\n",
      "2020-relook\n",
      "27-12-2007\n",
      "80s sci-fi\n",
      "60s sci-fi\n",
      "pg13_f-bomb\n",
      "made for 11-14-year-olds\n",
      "pg-13:intense sequences of sci-fi violence and action\n",
      "post-ww2\n",
      "2016 dvd-r\n",
      "pc-98\n",
      "instead of noodity why not say 3-piece birthday suits much more graphic\n",
      "3-d\n",
      "sr-71\n",
      "06 oscar nominated best movie - animation\n",
      "23-5-2007\n",
      "7-10-2007\n",
      "1-4-2007\n",
      "7-2007\n",
      "12-1-2007\n",
      "12-5-2007\n",
      "29-11-2006\n",
      "25-02-2007\n",
      "23-12-2007\n",
      "27-01-2007\n",
      "7-4-2007\n",
      "14-2-2007\n",
      "3-12-2006\n",
      "6-4-2007\n",
      "18-11-2006\n",
      "6-10-2007\n",
      "11-2-2008\n",
      "29--9-2007\n",
      "14-10-2007\n",
      "5-8-2007\n",
      "31-3-2007\n",
      "14-12-2006\n",
      "9-2007\n",
      "9-2-2007\n",
      "22-7-2007\n",
      "2-2-208\n",
      "2-2-2008\n",
      "10-2-2008\n",
      "19-8-2007\n",
      "17-1-2008\n",
      "5-7-2008\n",
      "7-year old girl\n",
      "220 foot free-fall\n",
      "fundamentalism; catch-22\n",
      "f-14\n",
      "surround sound - 6 ch\n",
      "13-year-old\n",
      "3-d animation\n",
      "retelling of 3 kings x-mas story\n",
      "pg-13:brief nudity\n",
      "pg-13:injury images\n",
      "pg-13:some strong language\n",
      "rated pg-13\n",
      "pg-13:drug content\n",
      "pg-13:war violence\n",
      "ex-imdb top 250\n",
      "pg-13:intense thematic elements\n",
      "pg-13:nudity/sensuality\n",
      "pg-13:sex-related humor\n",
      "pg-13:strong language\n",
      "pg-13:sexuality\n",
      "pg-13:some thematic elements\n",
      "16-year-old\n",
      "pg-13:language\n",
      "pg-13:violence\n",
      "pg-13:thematic issues involving teens\n",
      "pg-13:disturbing images\n",
      "pg-13:mature themes\n",
      "pg-13:some sexual references\n",
      "pg-13:intense violence and action\n",
      "pg-13:brief language\n",
      "pg-13:disturbing war images\n",
      "pg-13:brief drug use\n",
      "pg-13:mature thematic material involving sexual situations\n",
      "pg-13:drug use\n",
      "pg-13:sexual reference\n",
      "pg-13:sexual content\n",
      "pg-13:intense sequences of action\n",
      "pg-13:some drug use\n",
      "pg-13:some language\n",
      "pg-13:language including sexual references\n",
      "pg-13:mature thematic elements\n",
      "pg-13:a scene of sexuality\n",
      "pg-13:intense sequences of violence\n",
      "pg-13:some menace\n",
      "pg-13:a drug reference\n",
      "pg-13:crude and sexual humor\n",
      "pg-13:brief suggestive content\n",
      "pg-13:intense sci-fi action and violence\n",
      "pg-13:adventure violence\n",
      "pg-13:scary images\n",
      "pg-13:smoking\n",
      "pg-13:brief nude images\n",
      "pg-13:sexual situations including dialogue\n",
      "pg-13:thematic material\n",
      "pg-13:mature thematic material involving sexual content\n",
      "pg-13:action\n",
      "pg-13:alcohol use\n",
      "pg-13:mature thematic elements involving teen sexuality\n",
      "pg-13:some drug material\n",
      "pg-13:frightening images\n",
      "pg-13:some sequences of intense action violence\n",
      "pg-13:some sexual content including brief partial nudity\n",
      "pg-13:coarse humor\n",
      "pg-13:mild drug reference\n",
      "pg-13:some disturbing images and thematic material\n",
      "pg-13:disturbing images - all involving teens\n",
      "pg-13:intense violent thematic material\n",
      "pg-13:intense violent sequences throughout\n",
      "pg-13:extended sequences of intense fantasy action violence\n",
      "pg-13:some sensuality/nudity\n",
      "pg-13:sequences of intense sci-fi action and violence throughout\n",
      "pg-13:intense perilous sequences\n",
      "pg-13:some disturbing images\n",
      "pg-13:a suggestive situation\n",
      "pg-13:intense sequences of violence and action\n",
      "pg-13:some frightening images\n",
      "pg-13:historical smoking\n",
      "pg-13:some intense perilous action\n",
      "pg-13:gunplay and action throughout\n",
      "pg-13:mature thematic material\n",
      "pg-13: some disturbing images and thematic material\n",
      "pg-13:intense sequences of sci-fi action\n",
      "pg-13:some suggestive comments\n",
      "pg-13:violence and destruction\n",
      "pg-13:action violence\n",
      "pg-13:partial nudity\n",
      "pg-13:some suggestive content\n",
      "pg-13:thematic elements including violent content\n",
      "pg-13:some violence\n",
      "pg-13:suggestive material\n",
      "pg-13:some peril\n",
      "pg-13:extended sequences of sci-fi violence and action\n",
      "pg-13:intense war experience\n",
      "pg-13:some sci-fi action violence\n",
      "pg-13:some thematic content involving peril\n"
     ]
    }
   ],
   "source": [
    "def hasDigit(inputString):\n",
    "    return any(char.isdigit() for char in inputString)\n",
    "\n",
    "for x in tags_list_mod:\n",
    "    if hasDigit(x) and \"-\" in x:\n",
    "        print(x)"
   ]
  },
  {
   "attachments": {
    "image-2.png": {
     "image/png": "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"
    },
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Removemos as datas nesses formatos\n",
    "![image.png](attachment:image.png)\n",
    "\n",
    "E tags com esse formato:\n",
    "![image-2.png](attachment:image-2.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for index, row in data_tags_mod.iterrows():\n",
    "    tag = row.tag\n",
    "    \n",
    "    # Removendo parenteses e seus conteudos\n",
    "    corrected_tag = re.sub(r' \\([^)]*\\)', '', tag)\n",
    "    \n",
    "    # \"based\" é um palavra comum de estar nas tags, e vai evitar cair em outros ifs\n",
    "    if 'based' in corrected_tag:\n",
    "        data_tags_mod.loc[index, 'tag'] = corrected_tag\n",
    "        continue\n",
    "    # esse é para manter a tag sci-fi, e evita-la de ser removida no proximo if\n",
    "    if '-' in corrected_tag:\n",
    "        data_tags_mod.loc[index, 'tag'] = corrected_tag\n",
    "        continue\n",
    "    \n",
    "    if re.findall(r'\\b\\w{2}\\b', corrected_tag):\n",
    "        #todas as palavras de duas letras serão substituidas por NaN, que serão removidos depois\n",
    "        data_tags_mod.loc[index, 'tag'] = np.NaN\n",
    "    \n",
    "    elif re.findall(r'\\b\\w{1}\\b', corrected_tag):\n",
    "        #todas as palavras de uma letra serão substituidas por NaN, que serão removidos depois\n",
    "        data_tags_mod.loc[index, 'tag'] = np.NaN\n",
    "        \n",
    "    elif tag == corrected_tag:\n",
    "        #evitar substituição desnecessária\n",
    "        continue\n",
    "    \n",
    "    else:\n",
    "        data_tags_mod.loc[index, 'tag'] = corrected_tag\n",
    "        pass\n",
    "\n",
    "# remove todas as tags q são palavras de duas letras ou menos\n",
    "data_tags_mod = data_tags_mod.dropna()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Foram removidas todas as tags que continham palavras de duas ou uma letra, pois elas costumam não representar bem o filme. Exceto as \"sci-fi\" e as que contem \"based\", pois elas tem palavras de 2 letras e são importantes. (based on)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### O processamento de tags.csv foi finalizado acima, agora trabalharemos com os genomes."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Agora vamos adicionar o nome da tag (presente em genome-tags) à tabela de scores."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:36.208112Z",
     "start_time": "2021-06-24T15:46:34.342693Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Genome-Scores + Genome-Tags: \n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>tagId</th>\n",
       "      <th>tag</th>\n",
       "      <th>relevance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>0.02875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>0.04125</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>0.04675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>0.03425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>0.04300</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId  tagId  tag  relevance\n",
       "0        1      1  007    0.02875\n",
       "1        2      1  007    0.04125\n",
       "2        3      1  007    0.04675\n",
       "3        4      1  007    0.03425\n",
       "4        5      1  007    0.04300"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Fazendo o merge dos dois DataFrames\n",
    "data_scores = pd.merge(data_genomeScores, data_genomeTags)\n",
    "data_scores = data_scores[[\"movieId\", \"tagId\", \"tag\", \"relevance\"]]\n",
    "\n",
    "print(\"Genome-Scores + Genome-Tags: \")\n",
    "data_scores.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Como podemos observar, os valores de relevância para cada tag são valores contínuos. Para a abordagem Content-Based, é bom que tenhamos valores discretos para as características, desse modo, teremos então que agrupar os valores de relevância em categorias, podendo ser elas:\n",
    "\n",
    "#### \\begin{equation} 0 \\leq relevance < 0.25 = 1\\end{equation}\n",
    "#### \\begin{equation} 0.25 \\leq relevance < 0.5 = 2\\end{equation}\n",
    "#### \\begin{equation} 0.5 \\leq relevance < 0.75 = 3\\end{equation}\n",
    "#### \\begin{equation} 0.75 \\leq relevance  \\leq 1 = 4\\end{equation}\n",
    "\n",
    "\n",
    "#### Pode-se aumentar o numero de categorias para ter um modelo mais preciso"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:36.858258Z",
     "start_time": "2021-06-24T15:46:36.209112Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>tagId</th>\n",
       "      <th>tag</th>\n",
       "      <th>relevance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>007</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId  tagId  tag  relevance\n",
       "0        1      1  007        1.0\n",
       "1        2      1  007        1.0\n",
       "2        3      1  007        1.0\n",
       "3        4      1  007        1.0\n",
       "4        5      1  007        1.0\n",
       "5        6      1  007        1.0\n",
       "6        7      1  007        1.0\n",
       "7        8      1  007        1.0\n",
       "8        9      1  007        1.0\n",
       "9       10      1  007        4.0"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Vamos separar nos grupos de relevancia\n",
    "# Copiando\n",
    "data_tagRelevance = data_scores.copy()\n",
    "# Separando Grupo 1\n",
    "data_tagRelevance.loc[(data_tagRelevance.relevance < 0.25) & (data_tagRelevance.relevance >= 0), \"relevance\"] = 1\n",
    "# Separando Grupo 2\n",
    "data_tagRelevance.loc[(data_tagRelevance.relevance < 0.50) & (data_tagRelevance.relevance >= 0.25), \"relevance\"] = 2\n",
    "# Separando Grupo 3\n",
    "data_tagRelevance.loc[(data_tagRelevance.relevance < 0.75) & (data_tagRelevance.relevance >= 0.50), \"relevance\"] = 3\n",
    "# Separando Grupo 4\n",
    "data_tagRelevance.loc[(data_tagRelevance.relevance < 1.0) & (data_tagRelevance.relevance >= 0.75), \"relevance\"] = 4\n",
    "\n",
    "data_tagRelevance.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:37.264348Z",
     "start_time": "2021-06-24T15:46:36.859257Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "movieId\n",
       "1         1128\n",
       "57326     1128\n",
       "57368     1128\n",
       "57370     1128\n",
       "57393     1128\n",
       "          ... \n",
       "5107      1128\n",
       "5108      1128\n",
       "5109      1128\n",
       "5110      1128\n",
       "206499    1128\n",
       "Length: 13816, dtype: int64"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_tagRelevance.value_counts(\"movieId\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:37.660438Z",
     "start_time": "2021-06-24T15:46:37.265349Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "movieId\n",
       "1         1128\n",
       "57326     1128\n",
       "57368     1128\n",
       "57370     1128\n",
       "57393     1128\n",
       "          ... \n",
       "5107      1128\n",
       "5108      1128\n",
       "5109      1128\n",
       "5110      1128\n",
       "206499    1128\n",
       "Length: 13816, dtype: int64"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_tagRelevance.value_counts(\"movieId\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Transformaremos o DataFrame em um formato agrupado, com listas."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:43.004149Z",
     "start_time": "2021-06-24T15:46:37.661438Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>tag</th>\n",
       "      <th>relevance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 4.0, 3.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 3.0, 4.0, 4.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 3.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 4.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 4.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId  \\\n",
       "0        1   \n",
       "1        2   \n",
       "2        3   \n",
       "3        4   \n",
       "4        5   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   tag  \\\n",
       "0  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "1  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "2  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "3  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "4  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                             relevance  \n",
       "0  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 4.0, 3.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 3.0, 4.0, 4.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...  \n",
       "1  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 3.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 4.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 4.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, ...  \n",
       "2  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...  \n",
       "3  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, ...  \n",
       "4  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, ...  "
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Transformando o DataFrame, agrupando em listas com os valores de relevancia para cada Tag\n",
    "\n",
    "# Agrupando por movie Id, e fazendo uma lista das tags\n",
    "data_tagRelevanceGrouped = pd.DataFrame(data_tagRelevance.groupby(by=\"movieId\", axis=0)[\"tag\"].apply(list))\n",
    "# Agrupando por movie Id, e fazendo uma lista dos grupos de relevancia\n",
    "data_tagRelevanceGrouped[\"relevance\"] = pd.DataFrame(data_tagRelevance.groupby(by=\"movieId\", axis=0)[\"relevance\"].apply(list))\n",
    "# Resetando index\n",
    "data_tagRelevanceGrouped.reset_index(inplace=True)\n",
    "\n",
    "data_tagRelevanceGrouped.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:43.035156Z",
     "start_time": "2021-06-24T15:46:43.005150Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>tag</th>\n",
       "      <th>relevance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>13811</th>\n",
       "      <td>205072</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13812</th>\n",
       "      <td>205076</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13813</th>\n",
       "      <td>205383</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 2.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13814</th>\n",
       "      <td>205425</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 2.0, 2.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 2.0, 1.0, 2.0, 2.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13815</th>\n",
       "      <td>206499</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 2.0, 3.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 3.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       movieId  \\\n",
       "13811   205072   \n",
       "13812   205076   \n",
       "13813   205383   \n",
       "13814   205425   \n",
       "13815   206499   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       tag  \\\n",
       "13811  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "13812  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "13813  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "13814  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "13815  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 relevance  \n",
       "13811  [1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, ...  \n",
       "13812  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...  \n",
       "13813  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 2.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, ...  \n",
       "13814  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 2.0, 2.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 2.0, 1.0, 2.0, 2.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...  \n",
       "13815  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 2.0, 3.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 3.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, ...  "
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_tagRelevanceGrouped.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:44.061292Z",
     "start_time": "2021-06-24T15:46:43.036157Z"
    }
   },
   "outputs": [],
   "source": [
    "# Salvando as relevancias de tag agrupadas\n",
    "data_tagRelevanceGrouped.to_pickle(PATH_TO_TAG_RELEVANCE_GROUPED_CB_FILE)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Criando um dataframe igual ao anterior, porém com os valores de relevancia originais"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:49.244246Z",
     "start_time": "2021-06-24T15:46:44.062294Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>tag</th>\n",
       "      <th>relevance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[0.0287499999999999, 0.0237499999999999, 0.0625, 0.0757499999999999, 0.14075, 0.14675, 0.0635, 0.20375, 0.202, 0.03075, 0.58025, 0.1024999999999999, 0.2017499999999999, 0.007, 0.0245, 0.17275, 0.0165, 0.1039999999999999, 0.6625, 0.3007499999999999, 0.31675, 0.286, 0.0627499999999999, 0.0192499999999999, 0.05875, 0.0792499999999999, 0.1965, 0.09325, 0.89375, 0.67625, 0.03575, 0.23325, 0.3075, 0.0544999999999999, 0.0379999999999999, 0.212, 0.08, 0.0155, 0.0195, 0.0192499999999999, 0.015, 0.03,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[0.04125, 0.0404999999999999, 0.0627499999999999, 0.0827499999999999, 0.091, 0.06125, 0.0692499999999999, 0.0959999999999999, 0.0765, 0.0524999999999999, 0.088, 0.10275, 0.14525, 0.007, 0.0045, 0.0275, 0.0107499999999999, 0.15925, 0.64025, 0.3672499999999999, 0.5145, 0.4845, 0.1972499999999999, 0.24875, 0.0832499999999999, 0.1569999999999999, 0.1455, 0.05825, 0.976, 0.1267499999999999, 0.0105, 0.0735, 0.0707499999999999, 0.069, 0.10975, 0.2977499999999999, 0.062, 0.02, 0.0275, 0.017249999999...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[0.04675, 0.0554999999999999, 0.0292499999999999, 0.087, 0.0474999999999999, 0.04775, 0.0459999999999999, 0.14275, 0.0285, 0.03875, 0.0245, 0.0647499999999999, 0.081, 0.0055, 0.005, 0.03475, 0.0067499999999999, 0.11375, 0.16025, 0.14075, 0.252, 0.1937499999999999, 0.10425, 0.0167499999999999, 0.04875, 0.0645, 0.1317499999999999, 0.08625, 0.3215, 0.0954999999999999, 0.012, 0.03575, 0.05975, 0.0662499999999999, 0.0529999999999999, 0.2857499999999999, 0.05275, 0.029, 0.0364999999999999, 0.0175,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[0.03425, 0.0379999999999999, 0.0404999999999999, 0.031, 0.065, 0.03575, 0.029, 0.0865, 0.0319999999999999, 0.0314999999999999, 0.028, 0.03475, 0.0777499999999999, 0.0132499999999999, 0.0032499999999999, 0.0265, 0.0187499999999999, 0.1337499999999999, 0.147, 0.11025, 0.507, 0.46175, 0.1099999999999999, 0.017, 0.0745, 0.06075, 0.1252499999999999, 0.61775, 0.14875, 0.1315, 0.0135, 0.0404999999999999, 0.0524999999999999, 0.03725, 0.0454999999999999, 0.10175, 0.0964999999999999, 0.01224999999999...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[0.0429999999999999, 0.05325, 0.0379999999999999, 0.0409999999999999, 0.0539999999999999, 0.0672499999999999, 0.0277499999999999, 0.0765, 0.0215, 0.02975, 0.0277499999999999, 0.0329999999999999, 0.0504999999999999, 0.0082499999999999, 0.003, 0.0252499999999999, 0.018, 0.15475, 0.15575, 0.16825, 0.28925, 0.198, 0.089, 0.0225, 0.04425, 0.04775, 0.2175, 0.1129999999999999, 0.1634999999999999, 0.11875, 0.0115, 0.0349999999999999, 0.063, 0.0257499999999999, 0.10925, 0.1987499999999999, 0.03225, 0...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId  \\\n",
       "0        1   \n",
       "1        2   \n",
       "2        3   \n",
       "3        4   \n",
       "4        5   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   tag  \\\n",
       "0  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "1  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "2  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "3  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "4  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                             relevance  \n",
       "0  [0.0287499999999999, 0.0237499999999999, 0.0625, 0.0757499999999999, 0.14075, 0.14675, 0.0635, 0.20375, 0.202, 0.03075, 0.58025, 0.1024999999999999, 0.2017499999999999, 0.007, 0.0245, 0.17275, 0.0165, 0.1039999999999999, 0.6625, 0.3007499999999999, 0.31675, 0.286, 0.0627499999999999, 0.0192499999999999, 0.05875, 0.0792499999999999, 0.1965, 0.09325, 0.89375, 0.67625, 0.03575, 0.23325, 0.3075, 0.0544999999999999, 0.0379999999999999, 0.212, 0.08, 0.0155, 0.0195, 0.0192499999999999, 0.015, 0.03,...  \n",
       "1  [0.04125, 0.0404999999999999, 0.0627499999999999, 0.0827499999999999, 0.091, 0.06125, 0.0692499999999999, 0.0959999999999999, 0.0765, 0.0524999999999999, 0.088, 0.10275, 0.14525, 0.007, 0.0045, 0.0275, 0.0107499999999999, 0.15925, 0.64025, 0.3672499999999999, 0.5145, 0.4845, 0.1972499999999999, 0.24875, 0.0832499999999999, 0.1569999999999999, 0.1455, 0.05825, 0.976, 0.1267499999999999, 0.0105, 0.0735, 0.0707499999999999, 0.069, 0.10975, 0.2977499999999999, 0.062, 0.02, 0.0275, 0.017249999999...  \n",
       "2  [0.04675, 0.0554999999999999, 0.0292499999999999, 0.087, 0.0474999999999999, 0.04775, 0.0459999999999999, 0.14275, 0.0285, 0.03875, 0.0245, 0.0647499999999999, 0.081, 0.0055, 0.005, 0.03475, 0.0067499999999999, 0.11375, 0.16025, 0.14075, 0.252, 0.1937499999999999, 0.10425, 0.0167499999999999, 0.04875, 0.0645, 0.1317499999999999, 0.08625, 0.3215, 0.0954999999999999, 0.012, 0.03575, 0.05975, 0.0662499999999999, 0.0529999999999999, 0.2857499999999999, 0.05275, 0.029, 0.0364999999999999, 0.0175,...  \n",
       "3  [0.03425, 0.0379999999999999, 0.0404999999999999, 0.031, 0.065, 0.03575, 0.029, 0.0865, 0.0319999999999999, 0.0314999999999999, 0.028, 0.03475, 0.0777499999999999, 0.0132499999999999, 0.0032499999999999, 0.0265, 0.0187499999999999, 0.1337499999999999, 0.147, 0.11025, 0.507, 0.46175, 0.1099999999999999, 0.017, 0.0745, 0.06075, 0.1252499999999999, 0.61775, 0.14875, 0.1315, 0.0135, 0.0404999999999999, 0.0524999999999999, 0.03725, 0.0454999999999999, 0.10175, 0.0964999999999999, 0.01224999999999...  \n",
       "4  [0.0429999999999999, 0.05325, 0.0379999999999999, 0.0409999999999999, 0.0539999999999999, 0.0672499999999999, 0.0277499999999999, 0.0765, 0.0215, 0.02975, 0.0277499999999999, 0.0329999999999999, 0.0504999999999999, 0.0082499999999999, 0.003, 0.0252499999999999, 0.018, 0.15475, 0.15575, 0.16825, 0.28925, 0.198, 0.089, 0.0225, 0.04425, 0.04775, 0.2175, 0.1129999999999999, 0.1634999999999999, 0.11875, 0.0115, 0.0349999999999999, 0.063, 0.0257499999999999, 0.10925, 0.1987499999999999, 0.03225, 0...  "
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Transformando o DataFrame, agrupando em listas com os valores de relevancia para cada Tag\n",
    "\n",
    "# Agrupando por movie Id, e fazendo uma lista das tags\n",
    "data_tagGrouped = pd.DataFrame(data_scores.groupby(by=\"movieId\", axis=0)[\"tag\"].apply(list))\n",
    "# Agrupando por movie Id, e fazendo uma lista dos grupos de relevancia\n",
    "data_tagGrouped[\"relevance\"] = pd.DataFrame(data_scores.groupby(by=\"movieId\", axis=0)[\"relevance\"].apply(list))\n",
    "# Resetando index\n",
    "data_tagGrouped.reset_index(inplace=True)\n",
    "\n",
    "data_tagGrouped.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:49.276254Z",
     "start_time": "2021-06-24T15:46:49.245246Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>tag</th>\n",
       "      <th>relevance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>13811</th>\n",
       "      <td>205072</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[0.0205, 0.0177499999999999, 0.1139999999999999, 0.0364999999999999, 0.31225, 0.03675, 0.1069999999999999, 0.37925, 0.0172499999999999, 0.3635, 0.064, 0.0464999999999999, 0.23275, 0.0047499999999999, 0.0057499999999999, 0.0712499999999999, 0.0167499999999999, 0.15425, 0.22675, 0.1952499999999999, 0.38525, 0.2847499999999999, 0.0875, 0.0237499999999999, 0.16625, 0.05325, 0.23125, 0.0949999999999999, 0.24975, 0.353, 0.0197499999999999, 0.33425, 0.15475, 0.0155, 0.0082499999999999, 0.0424999999...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13812</th>\n",
       "      <td>205076</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[0.03825, 0.0314999999999999, 0.0319999999999999, 0.05325, 0.2085, 0.0705, 0.0662499999999999, 0.27825, 0.0095, 0.0275, 0.0305, 0.04225, 0.0459999999999999, 0.0067499999999999, 0.0802499999999999, 0.55225, 0.0185, 0.16575, 0.0964999999999999, 0.0777499999999999, 0.17925, 0.204, 0.0652499999999999, 0.0147499999999999, 0.0969999999999999, 0.24325, 0.10125, 0.1104999999999999, 0.124, 0.3482499999999999, 0.04725, 0.2534999999999999, 0.085, 0.028, 0.0257499999999999, 0.03075, 0.1215, 0.009, 0.004...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13813</th>\n",
       "      <td>205383</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[0.0409999999999999, 0.04025, 0.0275, 0.0785, 0.1975, 0.17825, 0.17125, 0.3047499999999999, 0.16825, 0.0484999999999999, 0.04275, 0.03675, 0.22075, 0.0115, 0.0042499999999999, 0.05025, 0.0202499999999999, 0.41825, 0.285, 0.27425, 0.52525, 0.23575, 0.087, 0.0145, 0.0705, 0.1375, 0.0979999999999999, 0.1134999999999999, 0.1887499999999999, 0.1945, 0.0122499999999999, 0.0554999999999999, 0.0459999999999999, 0.0685, 0.0964999999999999, 0.0974999999999999, 0.04625, 0.0275, 0.0499999999999999, 0.03...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13814</th>\n",
       "      <td>205425</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[0.04525, 0.04125, 0.0424999999999999, 0.0742499999999999, 0.1154999999999999, 0.1049999999999999, 0.0827499999999999, 0.1357499999999999, 0.16125, 0.05875, 0.03875, 0.04525, 0.2744999999999999, 0.0142499999999999, 0.0045, 0.06075, 0.0275, 0.34225, 0.26225, 0.16775, 0.4429999999999999, 0.3055, 0.1009999999999999, 0.0125, 0.2865, 0.1475, 0.27825, 0.148, 0.23975, 0.494, 0.0285, 0.36, 0.3255, 0.04625, 0.0275, 0.2584999999999999, 0.08975, 0.05975, 0.0334999999999999, 0.0102499999999999, 0.0125, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13815</th>\n",
       "      <td>206499</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[0.1004999999999999, 0.09325, 0.0222499999999999, 0.0454999999999999, 0.21125, 0.06125, 0.10425, 0.2975, 0.1245, 0.067, 0.0524999999999999, 0.0374999999999999, 0.1619999999999999, 0.0077499999999999, 0.0042499999999999, 0.0632499999999999, 0.0197499999999999, 0.50825, 0.4895, 0.25575, 0.501, 0.2574999999999999, 0.3904999999999999, 0.027, 0.09575, 0.10125, 0.06125, 0.3185, 0.2584999999999999, 0.27475, 0.0145, 0.071, 0.03725, 0.0974999999999999, 0.0717499999999999, 0.1347499999999999, 0.058499...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       movieId  \\\n",
       "13811   205072   \n",
       "13812   205076   \n",
       "13813   205383   \n",
       "13814   205425   \n",
       "13815   206499   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       tag  \\\n",
       "13811  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "13812  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "13813  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "13814  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "13815  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 relevance  \n",
       "13811  [0.0205, 0.0177499999999999, 0.1139999999999999, 0.0364999999999999, 0.31225, 0.03675, 0.1069999999999999, 0.37925, 0.0172499999999999, 0.3635, 0.064, 0.0464999999999999, 0.23275, 0.0047499999999999, 0.0057499999999999, 0.0712499999999999, 0.0167499999999999, 0.15425, 0.22675, 0.1952499999999999, 0.38525, 0.2847499999999999, 0.0875, 0.0237499999999999, 0.16625, 0.05325, 0.23125, 0.0949999999999999, 0.24975, 0.353, 0.0197499999999999, 0.33425, 0.15475, 0.0155, 0.0082499999999999, 0.0424999999...  \n",
       "13812  [0.03825, 0.0314999999999999, 0.0319999999999999, 0.05325, 0.2085, 0.0705, 0.0662499999999999, 0.27825, 0.0095, 0.0275, 0.0305, 0.04225, 0.0459999999999999, 0.0067499999999999, 0.0802499999999999, 0.55225, 0.0185, 0.16575, 0.0964999999999999, 0.0777499999999999, 0.17925, 0.204, 0.0652499999999999, 0.0147499999999999, 0.0969999999999999, 0.24325, 0.10125, 0.1104999999999999, 0.124, 0.3482499999999999, 0.04725, 0.2534999999999999, 0.085, 0.028, 0.0257499999999999, 0.03075, 0.1215, 0.009, 0.004...  \n",
       "13813  [0.0409999999999999, 0.04025, 0.0275, 0.0785, 0.1975, 0.17825, 0.17125, 0.3047499999999999, 0.16825, 0.0484999999999999, 0.04275, 0.03675, 0.22075, 0.0115, 0.0042499999999999, 0.05025, 0.0202499999999999, 0.41825, 0.285, 0.27425, 0.52525, 0.23575, 0.087, 0.0145, 0.0705, 0.1375, 0.0979999999999999, 0.1134999999999999, 0.1887499999999999, 0.1945, 0.0122499999999999, 0.0554999999999999, 0.0459999999999999, 0.0685, 0.0964999999999999, 0.0974999999999999, 0.04625, 0.0275, 0.0499999999999999, 0.03...  \n",
       "13814  [0.04525, 0.04125, 0.0424999999999999, 0.0742499999999999, 0.1154999999999999, 0.1049999999999999, 0.0827499999999999, 0.1357499999999999, 0.16125, 0.05875, 0.03875, 0.04525, 0.2744999999999999, 0.0142499999999999, 0.0045, 0.06075, 0.0275, 0.34225, 0.26225, 0.16775, 0.4429999999999999, 0.3055, 0.1009999999999999, 0.0125, 0.2865, 0.1475, 0.27825, 0.148, 0.23975, 0.494, 0.0285, 0.36, 0.3255, 0.04625, 0.0275, 0.2584999999999999, 0.08975, 0.05975, 0.0334999999999999, 0.0102499999999999, 0.0125, ...  \n",
       "13815  [0.1004999999999999, 0.09325, 0.0222499999999999, 0.0454999999999999, 0.21125, 0.06125, 0.10425, 0.2975, 0.1245, 0.067, 0.0524999999999999, 0.0374999999999999, 0.1619999999999999, 0.0077499999999999, 0.0042499999999999, 0.0632499999999999, 0.0197499999999999, 0.50825, 0.4895, 0.25575, 0.501, 0.2574999999999999, 0.3904999999999999, 0.027, 0.09575, 0.10125, 0.06125, 0.3185, 0.2584999999999999, 0.27475, 0.0145, 0.071, 0.03725, 0.0974999999999999, 0.0717499999999999, 0.1347499999999999, 0.058499...  "
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_tagGrouped.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:50.247871Z",
     "start_time": "2021-06-24T15:46:49.277254Z"
    }
   },
   "outputs": [],
   "source": [
    "# Salvando as relevancias de tags\n",
    "data_tagGrouped.to_pickle(PATH_TO_TAG_RELEVANCE_CB_FILE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:50.340892Z",
     "start_time": "2021-06-24T15:46:50.248871Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "      <th>year</th>\n",
       "      <th>average rating</th>\n",
       "      <th>rating count</th>\n",
       "      <th>weighted rating</th>\n",
       "      <th>tag</th>\n",
       "      <th>relevance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Toy Story</td>\n",
       "      <td>[Adventure, Animation, Children, Comedy, Fantasy]</td>\n",
       "      <td>1995</td>\n",
       "      <td>3.893708</td>\n",
       "      <td>57309</td>\n",
       "      <td>3.887824</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 4.0, 3.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 3.0, 4.0, 4.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>Jumanji</td>\n",
       "      <td>[Adventure, Children, Fantasy]</td>\n",
       "      <td>1995</td>\n",
       "      <td>3.251527</td>\n",
       "      <td>24228</td>\n",
       "      <td>3.248508</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 3.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 4.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 4.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Grumpier Old Men</td>\n",
       "      <td>[Comedy, Romance]</td>\n",
       "      <td>1995</td>\n",
       "      <td>3.142028</td>\n",
       "      <td>11804</td>\n",
       "      <td>3.139640</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>Waiting to Exhale</td>\n",
       "      <td>[Comedy, Drama, Romance]</td>\n",
       "      <td>1995</td>\n",
       "      <td>2.853547</td>\n",
       "      <td>2523</td>\n",
       "      <td>2.884188</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>Father of the Bride Part II</td>\n",
       "      <td>[Comedy]</td>\n",
       "      <td>1995</td>\n",
       "      <td>3.058434</td>\n",
       "      <td>11714</td>\n",
       "      <td>3.058875</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   movieId                        title  \\\n",
       "0        1                    Toy Story   \n",
       "1        2                      Jumanji   \n",
       "2        3             Grumpier Old Men   \n",
       "3        4            Waiting to Exhale   \n",
       "4        5  Father of the Bride Part II   \n",
       "\n",
       "                                              genres  year  average rating  \\\n",
       "0  [Adventure, Animation, Children, Comedy, Fantasy]  1995        3.893708   \n",
       "1                     [Adventure, Children, Fantasy]  1995        3.251527   \n",
       "2                                  [Comedy, Romance]  1995        3.142028   \n",
       "3                           [Comedy, Drama, Romance]  1995        2.853547   \n",
       "4                                           [Comedy]  1995        3.058434   \n",
       "\n",
       "   rating count  weighted rating  \\\n",
       "0         57309         3.887824   \n",
       "1         24228         3.248508   \n",
       "2         11804         3.139640   \n",
       "3          2523         2.884188   \n",
       "4         11714         3.058875   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   tag  \\\n",
       "0  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "1  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "2  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "3  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "4  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                             relevance  \n",
       "0  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 4.0, 3.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 3.0, 4.0, 4.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...  \n",
       "1  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 3.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 4.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 4.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, ...  \n",
       "2  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...  \n",
       "3  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, ...  \n",
       "4  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, ...  "
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#data_cb_final = pd.merge(data_cb, data_tagRelevanceGrouped)\n",
    "data_cb_final = pd.merge(data_cb, data_tagRelevanceGrouped)\n",
    "data_cb_final.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:50.371899Z",
     "start_time": "2021-06-24T15:46:50.341892Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>movieId</th>\n",
       "      <th>title</th>\n",
       "      <th>genres</th>\n",
       "      <th>year</th>\n",
       "      <th>average rating</th>\n",
       "      <th>rating count</th>\n",
       "      <th>weighted rating</th>\n",
       "      <th>tag</th>\n",
       "      <th>relevance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>13811</th>\n",
       "      <td>205072</td>\n",
       "      <td>Zombieland: Double Tap</td>\n",
       "      <td>[Action, Comedy, Horror]</td>\n",
       "      <td>2019</td>\n",
       "      <td>3.310345</td>\n",
       "      <td>58</td>\n",
       "      <td>3.100801</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13812</th>\n",
       "      <td>205076</td>\n",
       "      <td>Downton Abbey</td>\n",
       "      <td>[Drama]</td>\n",
       "      <td>2019</td>\n",
       "      <td>3.216981</td>\n",
       "      <td>53</td>\n",
       "      <td>3.087934</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13813</th>\n",
       "      <td>205383</td>\n",
       "      <td>El Camino: A Breaking Bad Movie</td>\n",
       "      <td>[Crime, Drama, Thriller]</td>\n",
       "      <td>2019</td>\n",
       "      <td>3.642857</td>\n",
       "      <td>252</td>\n",
       "      <td>3.287936</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 2.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13814</th>\n",
       "      <td>205425</td>\n",
       "      <td>Dave Chappelle: Sticks &amp; Stones</td>\n",
       "      <td>[Comedy]</td>\n",
       "      <td>2019</td>\n",
       "      <td>3.543478</td>\n",
       "      <td>69</td>\n",
       "      <td>3.138957</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 2.0, 2.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 2.0, 1.0, 2.0, 2.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13815</th>\n",
       "      <td>206499</td>\n",
       "      <td>Between Two Ferns: The Movie</td>\n",
       "      <td>[Comedy]</td>\n",
       "      <td>2019</td>\n",
       "      <td>3.055556</td>\n",
       "      <td>90</td>\n",
       "      <td>3.068544</td>\n",
       "      <td>[007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...</td>\n",
       "      <td>[1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 2.0, 3.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 3.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       movieId                            title                    genres  \\\n",
       "13811   205072           Zombieland: Double Tap  [Action, Comedy, Horror]   \n",
       "13812   205076                    Downton Abbey                   [Drama]   \n",
       "13813   205383  El Camino: A Breaking Bad Movie  [Crime, Drama, Thriller]   \n",
       "13814   205425  Dave Chappelle: Sticks & Stones                  [Comedy]   \n",
       "13815   206499     Between Two Ferns: The Movie                  [Comedy]   \n",
       "\n",
       "       year  average rating  rating count  weighted rating  \\\n",
       "13811  2019        3.310345            58         3.100801   \n",
       "13812  2019        3.216981            53         3.087934   \n",
       "13813  2019        3.642857           252         3.287936   \n",
       "13814  2019        3.543478            69         3.138957   \n",
       "13815  2019        3.055556            90         3.068544   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                       tag  \\\n",
       "13811  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "13812  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "13813  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "13814  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "13815  [007, 007 (series), 18th century, 1920s, 1930s, 1950s, 1960s, 1970s, 1980s, 19th century, 3d, 70mm, 80s, 9/11, aardman, aardman studios, abortion, absurd, action, action packed, adaptation, adapted from:book, adapted from:comic, adapted from:game, addiction, adolescence, adoption, adultery, adventure, affectionate, afi 100, afi 100 (laughs), afi 100 (movie quotes), africa, afterlife, aging, aids, airplane, airport, alaska, alcatraz, alcoholism, alien, alien invasion, aliens, allegory, almodo...   \n",
       "\n",
       "                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                 relevance  \n",
       "13811  [1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, ...  \n",
       "13812  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...  \n",
       "13813  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 2.0, 3.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, ...  \n",
       "13814  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 2.0, 2.0, 1.0, 1.0, 2.0, 1.0, 2.0, 1.0, 1.0, 2.0, 1.0, 2.0, 2.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, ...  \n",
       "13815  [1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 3.0, 2.0, 2.0, 3.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 3.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 1.0, 2.0, ...  "
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_cb_final.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-24T15:46:51.669073Z",
     "start_time": "2021-06-24T15:46:50.372899Z"
    }
   },
   "outputs": [],
   "source": [
    "# Salvando em .pickle o arquivo final\n",
    "data_cb_final.to_pickle(PATH_TO_FULL_CB_FILE)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# COMO SALVAR MANTENDO AS ESTRUTURAS E TIPOS DE DADOS\n",
    "## Save to pickle\n",
    "df.to_pickle('pickle-file.pkl')\n",
    "## Save with compression\n",
    "df.to_pickle('pickle-file.pkl.gz', compression='gzip')\n",
    "\n",
    "## Load pickle from disk\n",
    "df = pd.read_pickle('pickle-file.pkl')   # or...\n",
    "df = pd.read_pickle('pickle-file.pkl.gz', compression='gzip')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T19:48:16.552044Z",
     "start_time": "2021-05-27T19:48:13.123Z"
    }
   },
   "outputs": [],
   "source": [
    "# Plotar histograma da distribuição de notas\n",
    "plt.hist(averageRating_df['average rating'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T19:48:16.553045Z",
     "start_time": "2021-05-27T19:48:13.125Z"
    }
   },
   "outputs": [],
   "source": [
    "# Plotando histrograma da distribuição de notas com peso\n",
    "plt.hist(averageRating_df['weighted rating'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nota-se a grande concentração da quantidade de votos no intervalo de 0 à 18000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T19:48:16.554046Z",
     "start_time": "2021-05-27T19:48:13.129Z"
    }
   },
   "outputs": [],
   "source": [
    "sns.lmplot(x=\"number of ratings\", y=\"average rating\", data=averageRating_df)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Recuperar dados dos filmes do IMDb utilizando o imdbpy, para incrementar ainda mais os dados sobre os filmes.\n",
    "\n",
    "Cada filme registrado no IMDb tem um ID associado a ele, e o arquivo \"links.csv\", disponibilizado pelo MovieLens, trás todos os IDs relativos ao IMDb, o que irá nos auxiliar a recuperar os dados de cada filme."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T19:48:16.558045Z",
     "start_time": "2021-05-27T19:48:13.147Z"
    }
   },
   "outputs": [],
   "source": [
    "movies[0:2].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-05-27T19:48:16.559046Z",
     "start_time": "2021-05-27T19:48:13.149Z"
    }
   },
   "outputs": [],
   "source": [
    "# Criando objeto que irá fazer o web scraping\n",
    "imdb = IMDb()\n",
    "# Como a função do imdbpy \"get_movie\" utiliza os IDs dos filmes como parâmetro, iremos utilizar o DataFrame links\n",
    "# Criando o dataframe que irá guardar todas essas informações (pegaremos elenco (cast), diretores (directors), produtores(producers))\n",
    "\n",
    "# Criando o dicionario que irá armazenar as informações e posteriormente se tornará um DataFrame\n",
    "infos = {}\n",
    "\n",
    "# Pegando titulos do DataFrame \"movies\", e para cada titulo, criamos uma chave no dicionário, que irá ser da seguinte forma\n",
    "# {\"titulo\": [cast, directors, producers]}, onde cast, directors e producers são listas\n",
    "\n",
    "for row in movies[0:10].itertuples():\n",
    "    infos[row.movieId] = [row.title, row.genres, row.year, [], [], []]\n",
    "\n",
    "# Para cada linha no dataFrame \"links\" pesquisamos o filme no Imdb, recuperamos as informações extras e colocamos dentro do dict\n",
    "for row in links[0:10].itertuples():\n",
    "    movie = imdb.get_movie(row.imdbId, info=['main'])\n",
    "    \n",
    "    \n",
    "    for actor in movie['cast']:\n",
    "        # posição 3 da lista armazenada no dicionário é onde fica o cast\n",
    "        infos[row.movieId][3].append(actor['name'])\n",
    "    \n",
    "    for director in movie['directors']:\n",
    "        # posição 4 da lista armazenada no dicionário é onde fica os directors\n",
    "        infos[row.movieId][4].append(director['name'])\n",
    "        \n",
    "    for producer in movie['producers']:\n",
    "        # posição 5 da lista armazenada no dicionário é onde fica os producers\n",
    "        infos[row.movieId][5].append([producer['name']])\n",
    "    \n",
    "\n",
    "moviesWithMoreInfo = pd.DataFrame.from_dict(infos, orient='index', columns=['title', 'year','genres', 'cast', 'directors', 'producers'])\n",
    "\n",
    "moviesWithMoreInfo.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3 - Sistema de recomendação CF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-15T17:01:00.228668Z",
     "start_time": "2021-06-15T17:01:00.219666Z"
    }
   },
   "outputs": [],
   "source": [
    "# Importando bibliotecas para o sistema de recomendação\n",
    "from scipy.sparse import csr_matrix\n",
    "from sklearn.neighbors import NearestNeighbors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-15T17:01:00.784725Z",
     "start_time": "2021-06-15T17:01:00.402723Z"
    }
   },
   "outputs": [],
   "source": [
    "# Transformando a nossa matriz em uma mais eficiente utilizando o método csr (compressed sparse row matrix)\n",
    "user_movie_matrix_csr = csr_matrix(user_movie_matrix.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-15T17:01:00.800269Z",
     "start_time": "2021-06-15T17:01:00.785726Z"
    }
   },
   "outputs": [],
   "source": [
    "user_movie_matrix_csr.getrow(0).todense()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-15T17:01:00.893301Z",
     "start_time": "2021-06-15T17:01:00.880298Z"
    }
   },
   "outputs": [],
   "source": [
    "# Criando o modelo knn\n",
    "knn = NearestNeighbors(n_neighbors=5, algorithm='auto', metric='cosine') # temos que mexer nos parâmetros posteriormente\n",
    "\n",
    "knn.fit(user_movie_matrix_csr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-15T17:01:01.894071Z",
     "start_time": "2021-06-15T17:01:01.864063Z"
    }
   },
   "outputs": [],
   "source": [
    "movies_titles_list = []\n",
    "\n",
    "for title in data_movies[\"title\"]:\n",
    "    movies_titles_list.append(title)\n",
    "    \n",
    "# criando função que gera recomendações basedo em um filme\n",
    "def get_recomendations(movie_name):  \n",
    "    distances, suggestions = knn.kneighbors(user_movie_matrix.loc[movie_name].values.reshape(1, -1), n_neighbors=5)\n",
    "\n",
    "    for i in range(0, len(distances.flatten())):\n",
    "        if(i == 0):\n",
    "            print('Recomendações para {0}: \\n'.format(movie_name))\n",
    "        else:\n",
    "            print('{0}: {1}, com distância de {2}:'.format(i, user_movie_matrix.index[suggestions.flatten()[i]], distances.flatten()[i]))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-15T17:01:04.781677Z",
     "start_time": "2021-06-15T17:01:04.767673Z"
    }
   },
   "outputs": [],
   "source": [
    "# definindo função para pesquisar o nome correto do filme que está na matriz - lembrando que nem todo filme que está\n",
    "# no dataset está na matriz, pois precisamos usar apenas uma porção\n",
    "\n",
    "def search_movies_matrix(search_word):\n",
    "    return user_movie_matrix[user_movie_matrix.index.str.contains(search_word, flags=re.IGNORECASE)]\n",
    "\n",
    "# Função para pesquisar um filme dentro do dataset\n",
    "def search_movies(search_word):\n",
    "    return data_movies[data_movies.title.str.contains(search_word, flags=re.IGNORECASE)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-15T17:01:04.968223Z",
     "start_time": "2021-06-15T17:01:04.929214Z"
    }
   },
   "outputs": [],
   "source": [
    "# Procurando filmes geral\n",
    "search_movies(\"Ace Ventura: When Nature Calls\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-15T17:01:05.474337Z",
     "start_time": "2021-06-15T17:01:05.440330Z"
    }
   },
   "outputs": [],
   "source": [
    "# Procurando filmes na matriz\n",
    "search_movies_matrix(\"Ace \")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-15T17:01:06.924173Z",
     "start_time": "2021-06-15T17:01:06.893166Z"
    }
   },
   "outputs": [],
   "source": [
    "# Observando acima, queremos The Godfather - \"Godfather, The\"\n",
    "get_recomendations(\"Ace Ventura: When Nature Calls\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2021-06-15T17:01:56.275226Z",
     "start_time": "2021-06-15T17:01:56.256222Z"
    }
   },
   "outputs": [],
   "source": [
    "data_movies.loc[data_movies[\"title\"] == \"Braveheart\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}