text_preprocessing.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Basic Imports\n",
    "import os, warnings, openpyxl\n",
    "import nltk\n",
    "import numpy as np \n",
    "import pandas as pd \n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt \n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from collections import defaultdict\n",
    "\n",
    "# Data preparation and text-preprocessing\n",
    "import inflect, contractions, re, string, unicodedata, spacy\n",
    "from bs4 import BeautifulSoup\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.model_selection import train_test_split\n",
    "from nltk import word_tokenize, sent_tokenize\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.corpus import wordnet as wn\n",
    "from nltk.stem.wordnet import WordNetLemmatizer\n",
    "from nltk.stem import LancasterStemmer\n",
    "# from nltk.stem import WordNetLemmatizerd\n",
    "\n",
    "data_dir = os.path.join(os.getcwd(), \"Dataset\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nltk.download('all', quiet=True)\n",
    "nltk.download('wordnet', quiet=True)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training Datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "#3 Datasets\n",
    "tweets_df = pd.read_csv(os.path.join(data_dir,'Tweets_Emotions.csv'))\n",
    "emotions_df = pd.read_csv(os.path.join(data_dir, 'Emotions_from_text.csv'))\n",
    "sentiment_df = pd.read_csv(os.path.join(data_dir,\"Sentiment_140.csv\"), names=[\"sentiment\", \"id\", 'date',\"query\", \"user\", \"content\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tweets dataset\n",
      " tweet_id     0\n",
      "sentiment    0\n",
      "content      0\n",
      "dtype: int64 \n",
      "\n",
      "Emotions dataset\n",
      " Text       0\n",
      "Emotion    0\n",
      "dtype: int64 \n",
      "\n",
      "Sentiment\n",
      " sentiment    0\n",
      "id           0\n",
      "date         0\n",
      "query        0\n",
      "user         0\n",
      "content      0\n",
      "dtype: int64 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "#Check for null values\n",
    "print(\"Tweets dataset\\n\", tweets_df.isnull().sum(),\"\\n\")\n",
    "print(\"Emotions dataset\\n\", emotions_df.isnull().sum(),\"\\n\")\n",
    "print(\"Sentiment\\n\", sentiment_df.isnull().sum(),\"\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['empty' 'sadness' 'enthusiasm' 'neutral' 'worry' 'surprise' 'love' 'fun'\n",
      " 'hate' 'happiness' 'boredom' 'relief' 'anger'] \n",
      "\n",
      "['sadness' 'anger' 'love' 'surprise' 'fear' 'happy'] \n",
      "\n",
      "['positive' 'negative' 'neutral'] \n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(tweets_df[\"sentiment\"].unique(),\"\\n\")\n",
    "print(emotions_df[\"Emotion\"].unique(),\"\\n\")\n",
    "def sentiments_df3(val):\n",
    "    if val==0:\n",
    "        return 'negative'\n",
    "    elif val == 2:\n",
    "        return 'neutral'\n",
    "    elif val==4:\n",
    "        return 'positive'\n",
    "\n",
    "sentiment_df[\"sentiment\"] = sentiment_df[\"sentiment\"].apply(sentiments_df3)\n",
    "print(sentiment_df[\"sentiment\"].unique(),\"\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Removing unwanted columns\n",
    "tweets_df.drop([\"tweet_id\"], axis=1, inplace=True)\n",
    "tweets_df.drop(tweets_df.loc[tweets_df['sentiment']=='surprise'].index, inplace=True)\n",
    "emotions_df.drop(emotions_df.loc[emotions_df['Emotion']=='surprise'].index, inplace=True)\n",
    "sentiment_df.drop([\"id\", \"date\", \"query\", \"user\"], axis=1, inplace=True)\n",
    "\n",
    "#Standardizing the sentiment column naming\n",
    "emotions_df.rename(columns={\"Emotion\":\"sentiment\"}, inplace=True)\n",
    "emotions_df.rename(columns={\"Text\":\"content\"}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Combining all the datasets\n",
    "combined_df = pd.concat([tweets_df, emotions_df, sentiment_df], ignore_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>sentiment</th>\n",
       "      <th>content</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>empty</td>\n",
       "      <td>@tiffanylue i know  i was listenin to bad habi...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>sadness</td>\n",
       "      <td>Layin n bed with a headache  ughhhh...waitin o...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>sadness</td>\n",
       "      <td>Funeral ceremony...gloomy friday...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>enthusiasm</td>\n",
       "      <td>wants to hang out with friends SOON!</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>neutral</td>\n",
       "      <td>@dannycastillo We want to trade with someone w...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    sentiment                                            content\n",
       "0       empty  @tiffanylue i know  i was listenin to bad habi...\n",
       "1     sadness  Layin n bed with a headache  ughhhh...waitin o...\n",
       "2     sadness                Funeral ceremony...gloomy friday...\n",
       "3  enthusiasm               wants to hang out with friends SOON!\n",
       "4     neutral  @dannycastillo We want to trade with someone w..."
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combined_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['negative', 'positive'], dtype=object)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Classify the sentiment into positive and negative\n",
    "def generalise_sentiment(text):\n",
    "    if text=='love' or text=='positive' or text=='enthusiasm' or text =='surprise' or text=='fun 'or text=='happy' or text == 'happiness' or text=='relief':\n",
    "        return 'positive'\n",
    "    else:\n",
    "        return 'negative'\n",
    "\n",
    "combined_df[\"sentiment\"] = combined_df[\"sentiment\"].apply(generalise_sentiment)\n",
    "combined_df[\"sentiment\"].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "negative    38703\n",
       "positive    20188\n",
       "Name: sentiment, dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "combined_df[\"sentiment\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure()\n",
    "fig.patch.set_facecolor('white')\n",
    "plt.title(\"Sentiment Distribution\")\n",
    "plt.pie(combined_df[\"sentiment\"].value_counts(), labels=combined_df[\"sentiment\"].unique(), autopct='%1.1f%%')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "combined_df[\"char_len\"] = combined_df[\"content\"].apply(lambda x: len(x))\n",
    "combined_df[\"token_len\"] = combined_df[\"content\"].apply(lambda x: len(x.split(\" \")))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\ezeki\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\seaborn\\distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
      "  warnings.warn(msg, FutureWarning)\n",
      "c:\\Users\\ezeki\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\seaborn\\distributions.py:2619: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).\n",
      "  warnings.warn(msg, FutureWarning)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2)  = plt.subplots(nrows=1, ncols=2, figsize=(12,6))\n",
    "sns.distplot(combined_df['char_len'], ax=ax1)\n",
    "sns.distplot(combined_df['token_len'], ax=ax2)\n",
    "ax1.set_title('Number of characters in the tweet')\n",
    "ax2.set_title('Number of token(words) in the tweet')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "mean_df = combined_df.groupby(\"sentiment\").agg({\"char_len\":\"mean\", \"token_len\":\"mean\"})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\ezeki\\AppData\\Local\\Temp\\ipykernel_22832\\2646245128.py:6: UserWarning: FixedFormatter should only be used together with FixedLocator\n",
      "  ax1.set_xticklabels(mean_df.index, rotation = 45)\n",
      "C:\\Users\\ezeki\\AppData\\Local\\Temp\\ipykernel_22832\\2646245128.py:7: UserWarning: FixedFormatter should only be used together with FixedLocator\n",
      "  ax2.set_xticklabels(mean_df.index, rotation = 45)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1152x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2)  = plt.subplots(nrows=1, ncols=2, figsize=(16,6))\n",
    "ax1.bar(mean_df.index, mean_df['char_len'])\n",
    "ax2.bar(mean_df.index, mean_df['token_len'], color='green')\n",
    "ax1.set_title('Mean number of characters')\n",
    "ax2.set_title('Mean number of token(words)')\n",
    "ax1.set_xticklabels(mean_df.index, rotation = 45)\n",
    "ax2.set_xticklabels(mean_df.index, rotation = 45)\n",
    "plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Text Pre-Processing Function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "def text_preprocessing_platform(df, text_col, remove_stopwords=True):\n",
    "    ## Define functions for individual steps\n",
    "    # First function is used to denoise text\n",
    "    def denoise_text(text):\n",
    "        # Strip html if any. For ex. removing <html>, <p> tags\n",
    "        soup = BeautifulSoup(text, \"html.parser\")\n",
    "        text = soup.get_text()\n",
    "        # Replace contractions in the text. For ex. didn't -> did not\n",
    "        text = contractions.fix(text)\n",
    "        return text\n",
    "    \n",
    "    def remove_non_ascii(words):\n",
    "        \"\"\"Remove non-ASCII characters from list of tokenized words\"\"\"\n",
    "        new_words = []\n",
    "        for word in words:\n",
    "            new_word = unicodedata.normalize('NFKD', word).encode('ascii', 'ignore').decode('utf-8', 'ignore')\n",
    "            new_words.append(new_word)\n",
    "        return new_words\n",
    "    \n",
    "    def to_lowercase(words):\n",
    "        \"\"\"Convert all characters to lowercase from list of tokenized words\"\"\"\n",
    "        new_words = []\n",
    "        for word in words:\n",
    "            new_word = word.lower()\n",
    "            new_words.append(new_word)\n",
    "        return new_words\n",
    "    \n",
    "    def remove_punctuation(words):\n",
    "        \"\"\"Remove punctuation from list of tokenized words\"\"\"\n",
    "        new_words = []\n",
    "        for word in words:\n",
    "            new_word = re.sub(r'[^\\w\\s]', '', word)\n",
    "            if new_word != '':\n",
    "                new_words.append(new_word)\n",
    "        return new_words\n",
    "    \n",
    "    def replace_numbers(words):\n",
    "        \"\"\"Replace all interger occurrences in list of tokenized words with textual representation\"\"\"\n",
    "        p = inflect.engine()\n",
    "        new_words = []\n",
    "        for word in words:\n",
    "            if word.isdigit():\n",
    "                new_word = p.number_to_words(word)\n",
    "                new_words.append(new_word)\n",
    "            else:\n",
    "                new_words.append(word)\n",
    "        return new_words\n",
    "    \n",
    "    def remove_stopwords(words):\n",
    "        \"\"\"Remove stop words from list of tokenized words\"\"\"\n",
    "        new_words = []\n",
    "        for word in words:\n",
    "            if word not in stopwords.words('english'):\n",
    "                new_words.append(word)\n",
    "        return new_words\n",
    "    \n",
    "    def stem_words(words):\n",
    "        \"\"\"Stem words in list of tokenized words\"\"\"\n",
    "        stemmer = LancasterStemmer()\n",
    "        stems = []\n",
    "        for word in words:\n",
    "            stem = stemmer.stem(word)\n",
    "            stems.append(stem)\n",
    "        return stems\n",
    "    \n",
    "    def lemmatize_verbs(words):\n",
    "        \"\"\"Lemmatize verbs in list of tokenized words\"\"\"\n",
    "        lemmatizer = WordNetLemmatizer()\n",
    "        lemmas = []\n",
    "        for word in words:\n",
    "            lemma = lemmatizer.lemmatize(word, pos='v')\n",
    "            lemmas.append(lemma)\n",
    "        return lemmas\n",
    "    \n",
    "    ### A wrap-up function for normalization\n",
    "    def normalize_text(words, remove_stopwords):\n",
    "        words = remove_non_ascii(words)\n",
    "        words = to_lowercase(words)\n",
    "        words = remove_punctuation(words)\n",
    "        words = replace_numbers(words)\n",
    "        if remove_stopwords:\n",
    "            words = remove_stopwords(words)\n",
    "        #words = stem_words(words)\n",
    "        words = lemmatize_verbs(words)\n",
    "        return words\n",
    "\n",
    "    # Tokenize tweet into words\n",
    "    def tokenize(text):\n",
    "        return nltk.word_tokenize(text)\n",
    "\n",
    "    # A overall wrap-up function\n",
    "    def text_prepare(text):\n",
    "        text = denoise_text(text)\n",
    "        text = ' '.join([x for x in normalize_text(tokenize(text), remove_stopwords)])\n",
    "        return text\n",
    "    \n",
    "    # run every-step\n",
    "    df[text_col] = [text_prepare(x) for x in df[text_col]]\n",
    "\n",
    "    # return processed df\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Before Text Preprocessing\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>content</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>52104</th>\n",
       "      <td>i went by on wednesday feeling slightly regret...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>690</th>\n",
       "      <td>@textdrivebys i dont even have a WII! b likes ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33683</th>\n",
       "      <td>@mini_ritz thnx sweetie  how r u this evening? xo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17853</th>\n",
       "      <td>@MariaV_ST vaca!! buuu sigo en el work</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45122</th>\n",
       "      <td>i feel fond toward though they may not realize it</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                 content\n",
       "52104  i went by on wednesday feeling slightly regret...\n",
       "690    @textdrivebys i dont even have a WII! b likes ...\n",
       "33683  @mini_ritz thnx sweetie  how r u this evening? xo\n",
       "17853             @MariaV_ST vaca!! buuu sigo en el work\n",
       "45122  i feel fond toward though they may not realize it"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\ezeki\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\bs4\\__init__.py:435: MarkupResemblesLocatorWarning: The input looks more like a filename than markup. You may want to open this file and pass the filehandle into Beautiful Soup.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "After Text Preprocessing\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>content</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>365</th>\n",
       "      <td>count hours lose sunshine weekend</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33921</th>\n",
       "      <td>lovetogivenuts yet save page read travel</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39825</th>\n",
       "      <td>say feel like verge messy</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46053</th>\n",
       "      <td>feel weird yuuki talk girls</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29285</th>\n",
       "      <td>eat salad rember comment new myspace picture plz</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                content\n",
       "365                   count hours lose sunshine weekend\n",
       "33921          lovetogivenuts yet save page read travel\n",
       "39825                         say feel like verge messy\n",
       "46053                       feel weird yuuki talk girls\n",
       "29285  eat salad rember comment new myspace picture plz"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Running text preprocessing on the combined dataset\n",
    "print(\"Before Text Preprocessing\")\n",
    "display(combined_df.sample(5)[['content']])\n",
    "\n",
    "processed_df = text_preprocessing_platform(combined_df, 'content', remove_stopwords=False)\n",
    "\n",
    "print(\"After Text Preprocessing\")\n",
    "display(processed_df.sample(5)[['content']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save the processed dataframe\n",
    "processed_df.to_csv(os.path.join(data_dir,\"processed_df.csv\"), index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "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>sentiment</th>\n",
       "      <th>content</th>\n",
       "      <th>char_len</th>\n",
       "      <th>token_len</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>50265</th>\n",
       "      <td>positive</td>\n",
       "      <td>feel privilege experience lovely place able vi...</td>\n",
       "      <td>115</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20729</th>\n",
       "      <td>positive</td>\n",
       "      <td>nirajgoa oh love vodafone series ads best pict...</td>\n",
       "      <td>94</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30208</th>\n",
       "      <td>positive</td>\n",
       "      <td>sicilyyoder yet eat couple love reeces hard ge...</td>\n",
       "      <td>120</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40945</th>\n",
       "      <td>positive</td>\n",
       "      <td>uneasy feel stupidly talented eagle mainly goo...</td>\n",
       "      <td>147</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15422</th>\n",
       "      <td>positive</td>\n",
       "      <td>johnreppion awww let would better next 30th xd</td>\n",
       "      <td>84</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      sentiment                                            content  char_len  \\\n",
       "50265  positive  feel privilege experience lovely place able vi...       115   \n",
       "20729  positive  nirajgoa oh love vodafone series ads best pict...        94   \n",
       "30208  positive  sicilyyoder yet eat couple love reeces hard ge...       120   \n",
       "40945  positive  uneasy feel stupidly talented eagle mainly goo...       147   \n",
       "15422  positive     johnreppion awww let would better next 30th xd        84   \n",
       "\n",
       "       token_len  \n",
       "50265         21  \n",
       "20729         16  \n",
       "30208         27  \n",
       "40945         26  \n",
       "15422         18  "
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "processed_df.sample(5)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Can read processed text directly into dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "processed_df = pd.read_csv(os.path.join(data_dir,\"processed_df.csv\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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>sentiment</th>\n",
       "      <th>content</th>\n",
       "      <th>char_len</th>\n",
       "      <th>token_len</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>53826</th>\n",
       "      <td>negative</td>\n",
       "      <td>feel greedy right</td>\n",
       "      <td>31</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30832</th>\n",
       "      <td>negative</td>\n",
       "      <td>lindentreephoto unless mean two hours week use...</td>\n",
       "      <td>103</td>\n",
       "      <td>19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2569</th>\n",
       "      <td>negative</td>\n",
       "      <td>dean_b drive home think break laws tooooooo ma...</td>\n",
       "      <td>123</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12388</th>\n",
       "      <td>negative</td>\n",
       "      <td>another two hours work ugh go slow sore</td>\n",
       "      <td>63</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20063</th>\n",
       "      <td>negative</td>\n",
       "      <td>challiyan wait till england wc t20</td>\n",
       "      <td>35</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      sentiment                                            content  char_len  \\\n",
       "53826  negative                                  feel greedy right        31   \n",
       "30832  negative  lindentreephoto unless mean two hours week use...       103   \n",
       "2569   negative  dean_b drive home think break laws tooooooo ma...       123   \n",
       "12388  negative            another two hours work ugh go slow sore        63   \n",
       "20063  negative                 challiyan wait till england wc t20        35   \n",
       "\n",
       "       token_len  \n",
       "53826          6  \n",
       "30832         19  \n",
       "2569          26  \n",
       "12388         14  \n",
       "20063          6  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "processed_df.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.10.2 (tags/v3.10.2:a58ebcc, Jan 17 2022, 14:12:15) [MSC v.1929 64 bit (AMD64)]"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "54b5c02bac415170f398c1cb9c40d2712252a7d195ee8eb19ed147dda00b9728"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}