NaiveBayes_Yelp_30K_BM.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Dependencies\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import os\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from pyspark.sql import SparkSession\n",
    "from pyspark.ml.feature import Tokenizer, NGram\n",
    "from pyspark.sql.functions import length\n",
    "from pyspark.ml.feature import Tokenizer, StopWordsRemover, HashingTF, IDF, StringIndexer\n",
    "from pyspark.ml.feature import VectorAssembler\n",
    "from pyspark.ml.linalg import Vector\n",
    "from pyspark.ml import Pipeline\n",
    "from pyspark.ml.classification import NaiveBayes\n",
    "from pyspark.ml.evaluation import MulticlassClassificationEvaluator\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>review_id</th>\n",
       "      <th>user_id</th>\n",
       "      <th>business_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>text</th>\n",
       "      <th>date</th>\n",
       "      <th>review_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>\"Brick House Tavern + Tap\"</td>\n",
       "      <td>ZflgoahrEd1_H0tWSATjQw</td>\n",
       "      <td>KgQnRvsPSrJJTOCp-reYZA</td>\n",
       "      <td>PfOCPjBrlQAnz__NXj9h_w</td>\n",
       "      <td>4</td>\n",
       "      <td>Eating on couches or lounging in comfy recline...</td>\n",
       "      <td>2016-02-27</td>\n",
       "      <td>57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>\"Whole Foods Market\"</td>\n",
       "      <td>JjJyUhnLtX5aZWVJfSeLKA</td>\n",
       "      <td>KgQnRvsPSrJJTOCp-reYZA</td>\n",
       "      <td>NFv21m2Czqz2XM5G4kpbKg</td>\n",
       "      <td>3</td>\n",
       "      <td>Convienent easy parking, though I had to avoid...</td>\n",
       "      <td>2016-12-27</td>\n",
       "      <td>57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>\"Crave Cantina\"</td>\n",
       "      <td>hQu_gojfKPm8C0yovoszhQ</td>\n",
       "      <td>KgQnRvsPSrJJTOCp-reYZA</td>\n",
       "      <td>OFaMjL9nLF-A51KHNW8nfg</td>\n",
       "      <td>5</td>\n",
       "      <td>I usually go somewhere local twice before writ...</td>\n",
       "      <td>2017-07-01</td>\n",
       "      <td>57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>\"Newdle Bar\"</td>\n",
       "      <td>0n1w0w1VsRF4eqeqZHnX4g</td>\n",
       "      <td>KgQnRvsPSrJJTOCp-reYZA</td>\n",
       "      <td>XHK_1UM4CBZQ7RemDFr1mQ</td>\n",
       "      <td>3</td>\n",
       "      <td>Meh...don't come here if you are looking for a...</td>\n",
       "      <td>2016-08-11</td>\n",
       "      <td>57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>\"Cornerstone Market &amp; Deli\"</td>\n",
       "      <td>PAldTKppEliQBy3M5vxtlg</td>\n",
       "      <td>KgQnRvsPSrJJTOCp-reYZA</td>\n",
       "      <td>eZ423jsTh3g9QTt2w-UTyg</td>\n",
       "      <td>5</td>\n",
       "      <td>This place is a gem...I was so glad they reope...</td>\n",
       "      <td>2016-08-17</td>\n",
       "      <td>57</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          name               review_id  \\\n",
       "0   \"Brick House Tavern + Tap\"  ZflgoahrEd1_H0tWSATjQw   \n",
       "1         \"Whole Foods Market\"  JjJyUhnLtX5aZWVJfSeLKA   \n",
       "2              \"Crave Cantina\"  hQu_gojfKPm8C0yovoszhQ   \n",
       "3                 \"Newdle Bar\"  0n1w0w1VsRF4eqeqZHnX4g   \n",
       "4  \"Cornerstone Market & Deli\"  PAldTKppEliQBy3M5vxtlg   \n",
       "\n",
       "                  user_id             business_id  stars  \\\n",
       "0  KgQnRvsPSrJJTOCp-reYZA  PfOCPjBrlQAnz__NXj9h_w      4   \n",
       "1  KgQnRvsPSrJJTOCp-reYZA  NFv21m2Czqz2XM5G4kpbKg      3   \n",
       "2  KgQnRvsPSrJJTOCp-reYZA  OFaMjL9nLF-A51KHNW8nfg      5   \n",
       "3  KgQnRvsPSrJJTOCp-reYZA  XHK_1UM4CBZQ7RemDFr1mQ      3   \n",
       "4  KgQnRvsPSrJJTOCp-reYZA  eZ423jsTh3g9QTt2w-UTyg      5   \n",
       "\n",
       "                                                text        date  review_count  \n",
       "0  Eating on couches or lounging in comfy recline...  2016-02-27            57  \n",
       "1  Convienent easy parking, though I had to avoid...  2016-12-27            57  \n",
       "2  I usually go somewhere local twice before writ...  2017-07-01            57  \n",
       "3  Meh...don't come here if you are looking for a...  2016-08-11            57  \n",
       "4  This place is a gem...I was so glad they reope...  2016-08-17            57  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "file_name = \"reviews_users_50-100.csv\"\n",
    "yelp_review_df = pd.read_csv(file_name, encoding = \"ISO-8859-1\")\n",
    "yelp_review_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "short_yelp_review_df = yelp_review_df.iloc[:70000]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_id</th>\n",
       "      <th>text</th>\n",
       "      <th>stars</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ZflgoahrEd1_H0tWSATjQw</td>\n",
       "      <td>Eating on couches or lounging in comfy recline...</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>JjJyUhnLtX5aZWVJfSeLKA</td>\n",
       "      <td>Convienent easy parking, though I had to avoid...</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>hQu_gojfKPm8C0yovoszhQ</td>\n",
       "      <td>I usually go somewhere local twice before writ...</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0n1w0w1VsRF4eqeqZHnX4g</td>\n",
       "      <td>Meh...don't come here if you are looking for a...</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>PAldTKppEliQBy3M5vxtlg</td>\n",
       "      <td>This place is a gem...I was so glad they reope...</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                review_id                                               text  \\\n",
       "0  ZflgoahrEd1_H0tWSATjQw  Eating on couches or lounging in comfy recline...   \n",
       "1  JjJyUhnLtX5aZWVJfSeLKA  Convienent easy parking, though I had to avoid...   \n",
       "2  hQu_gojfKPm8C0yovoszhQ  I usually go somewhere local twice before writ...   \n",
       "3  0n1w0w1VsRF4eqeqZHnX4g  Meh...don't come here if you are looking for a...   \n",
       "4  PAldTKppEliQBy3M5vxtlg  This place is a gem...I was so glad they reope...   \n",
       "\n",
       "   stars  \n",
       "0      4  \n",
       "1      3  \n",
       "2      5  \n",
       "3      3  \n",
       "4      5  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_df = short_yelp_review_df[[\"review_id\", \"text\",\"stars\"]].reset_index()\n",
    "test_df = test_df[[\"review_id\",'text', 'stars']]\n",
    "test_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "spark = SparkSession.builder.appName('Yelp').getOrCreate()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "start_data = spark.createDataFrame(test_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+--------------------+--------------------+-----+------+\n",
      "|           review_id|                text|stars|length|\n",
      "+--------------------+--------------------+-----+------+\n",
      "|ZflgoahrEd1_H0tWS...|Eating on couches...|    4|   744|\n",
      "|JjJyUhnLtX5aZWVJf...|Convienent easy p...|    3|   668|\n",
      "|hQu_gojfKPm8C0yov...|I usually go some...|    5|  1019|\n",
      "|0n1w0w1VsRF4eqeqZ...|Meh...don't come ...|    3|   559|\n",
      "|PAldTKppEliQBy3M5...|This place is a g...|    5|   569|\n",
      "|r8ghBuwFFMybBGFcD...|Step back in time...|    4|   345|\n",
      "|PTldCrPlj6FbJnekN...|I have been twice...|    4|   695|\n",
      "|I1NMsaLQXLQRFEJwk...|This restaurant g...|    5|   639|\n",
      "|moFCqVwlzeUE4em5P...|We will be back h...|    5|   537|\n",
      "|oEa4Csf_LjOQHiwRa...|Came in due to re...|    3|   525|\n",
      "|otCspYf4Vzo97zAuo...|Always great, fas...|    4|   222|\n",
      "|n3DUEVDu2zbm4ewGS...|We have been here...|    4|   271|\n",
      "|fPbWkOn2hhkNu8JvZ...|The smell of cara...|    5|   597|\n",
      "|bVpzB-6wfMz_easvZ...|Preferred lunch v...|    3|   845|\n",
      "|aT_6C6KU64l87unF7...|Great burgers...h...|    5|   369|\n",
      "|3JTTBk-pg3mcWIyjy...|Wish I could give...|    4|  1164|\n",
      "|OiGZiZx2QE7Dgk3qS...|What is there to ...|    3|   407|\n",
      "|1dJDr6nyEdyDCjV0d...|The parking here ...|    3|   593|\n",
      "|0K57zqClL8x2LWn6-...|Pay attention to ...|    4|   276|\n",
      "|vLAg1X9pdNjF4ZOKf...|What a fun store!...|    5|   272|\n",
      "+--------------------+--------------------+-----+------+\n",
      "only showing top 20 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "\n",
    "data = start_data.withColumn('length', length(start_data['text']))\n",
    "\n",
    "data.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Create all the features to the data set\n",
    "\n",
    "stars_to_num = StringIndexer(inputCol='stars',outputCol='label')\n",
    "tokenizer = Tokenizer(inputCol=\"text\", outputCol=\"token_text\")\n",
    "ngram = NGram(n=2, inputCol=\"token_text\", outputCol=\"ngrams\")\n",
    "stopremove = StopWordsRemover(inputCol=\"ngrams\",outputCol='stop_tokens')\n",
    "hashingTF = HashingTF(inputCol='stop_tokens', outputCol='hash_token')\n",
    "idf = IDF(inputCol='hash_token', outputCol='idf_token')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Create feature vectors\n",
    "clean_up = VectorAssembler(inputCols=['idf_token', 'length'], outputCol='features')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data_prep_pipeline = Pipeline(stages=[stars_to_num, tokenizer, ngram, stopremove, hashingTF, idf, clean_up]) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Fit and transform the pipeline\n",
    "cleaner = data_prep_pipeline.fit(data)\n",
    "cleaned = cleaner.transform(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+-----+--------------------+\n",
      "|label|            features|\n",
      "+-----+--------------------+\n",
      "|  0.0|(262145,[303,3143...|\n",
      "|  2.0|(262145,[733,811,...|\n",
      "|  1.0|(262145,[258,429,...|\n",
      "|  2.0|(262145,[43,1479,...|\n",
      "|  1.0|(262145,[1634,399...|\n",
      "|  0.0|(262145,[7787,831...|\n",
      "|  0.0|(262145,[43,815,1...|\n",
      "|  1.0|(262145,[1447,369...|\n",
      "|  1.0|(262145,[2826,552...|\n",
      "|  2.0|(262145,[43,964,2...|\n",
      "|  0.0|(262145,[22553,33...|\n",
      "|  0.0|(262145,[5355,600...|\n",
      "|  1.0|(262145,[1447,437...|\n",
      "|  2.0|(262145,[8755,108...|\n",
      "|  1.0|(262145,[2925,898...|\n",
      "|  0.0|(262145,[83,1447,...|\n",
      "|  2.0|(262145,[1413,281...|\n",
      "|  2.0|(262145,[207,732,...|\n",
      "|  0.0|(262145,[5155,600...|\n",
      "|  1.0|(262145,[1481,314...|\n",
      "+-----+--------------------+\n",
      "only showing top 20 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "cleaned.select(['label', 'features']).show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Break data down into a training set and a testing set\n",
    "(training, testing) = cleaned.randomSplit([0.7, 0.3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Create a Naive Bayes model and fit training data\n",
    "nb = NaiveBayes(smoothing=1.0, modelType='multinomial')   \n",
    "star_predictor = nb.fit(training)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "+--------------------+--------------------+-----+------+-----+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+----------+\n",
      "|           review_id|                text|stars|length|label|          token_text|              ngrams|         stop_tokens|          hash_token|           idf_token|            features|       rawPrediction|         probability|prediction|\n",
      "+--------------------+--------------------+-----+------+-----+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+----------+\n",
      "|--VE-t7k_ykSu2mlK...|I work in Park Ce...|    4|   615|  0.0|[i, work, in, par...|[i work, work in,...|[i work, work in,...|(262144,[4865,600...|(262144,[4865,600...|(262145,[4865,600...|[-8922.0932401670...|[0.95620643550424...|       0.0|\n",
      "|--yzATieNipvZ7tVu...|This place caught...|    5|   412|  1.0|[this, place, cau...|[this place, plac...|[this place, plac...|(262144,[665,1447...|(262144,[665,1447...|(262145,[665,1447...|[-5982.1331885193...|[0.99935098638845...|       0.0|\n",
      "|-0HGoSW-aswXr2dlc...|amazing place to ...|    4|   443|  0.0|[amazing, place, ...|[amazing place, p...|[amazing place, p...|(262144,[3059,586...|(262144,[3059,586...|(262145,[3059,586...|[-6649.5822125214...|[8.54324188089170...|       2.0|\n",
      "|-0sjL8VTtLDLBBywR...|You can't ever go...|    4|   513|  0.0|[you, can't, ever...|[you can't, can't...|[you can't, can't...|(262144,[2152,426...|(262144,[2152,426...|(262145,[2152,426...|[-7508.7630573970...|[3.72213676095227...|       1.0|\n",
      "|-2ayGrxgMlRAQpix5...|I came here the o...|    4|   359|  0.0|[i, came, here, t...|[i came, came her...|[i came, came her...|(262144,[7412,756...|(262144,[7412,756...|(262145,[7412,756...|[-4983.3070159709...|[6.88796685855943...|       1.0|\n",
      "+--------------------+--------------------+-----+------+-----+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+--------------------+----------+\n",
      "only showing top 5 rows\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Tranform the model with the testing data\n",
    "test_results = star_predictor.transform(testing)\n",
    "test_results.show(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy of model at predicting reviews was: 0.4389972107129357\n"
     ]
    }
   ],
   "source": [
    "# Use the Class Evaluator for a cleaner description\n",
    "\n",
    "\n",
    "acc_eval = MulticlassClassificationEvaluator()\n",
    "acc = acc_eval.evaluate(test_results)\n",
    "print(f\"Accuracy of model at predicting reviews was: {acc}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>label</th>\n",
       "      <th>prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>--VE-t7k_ykSu2mlKpKwqA</td>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>--yzATieNipvZ7tVuTvNKw</td>\n",
       "      <td>5</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0HGoSW-aswXr2dlcqb9Dw</td>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0sjL8VTtLDLBBywRG6NJg</td>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-2ayGrxgMlRAQpix5RZtTA</td>\n",
       "      <td>4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                review_id  stars  label  prediction\n",
       "0  --VE-t7k_ykSu2mlKpKwqA      4    0.0         0.0\n",
       "1  --yzATieNipvZ7tVuTvNKw      5    1.0         0.0\n",
       "2  -0HGoSW-aswXr2dlcqb9Dw      4    0.0         2.0\n",
       "3  -0sjL8VTtLDLBBywRG6NJg      4    0.0         1.0\n",
       "4  -2ayGrxgMlRAQpix5RZtTA      4    0.0         1.0"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_results_pandas_df = test_results.select(\"review_id\",\"stars\", \"label\", \"prediction\").toPandas()\n",
    "test_results_pandas_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>label</th>\n",
       "      <th>prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [review_id, stars, label, prediction]\n",
       "Index: []"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_results_pandas_df[test_results_pandas_df[\"review_id\"] == 'NHuuLXPWt9d9JXmptz3zjg']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>review_id</th>\n",
       "      <th>stars</th>\n",
       "      <th>label</th>\n",
       "      <th>prediction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>153</th>\n",
       "      <td>0n1w0w1VsRF4eqeqZHnX4g</td>\n",
       "      <td>3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  review_id  stars  label  prediction\n",
       "153  0n1w0w1VsRF4eqeqZHnX4g      3    2.0         2.0"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_results_pandas_df[test_results_pandas_df[\"review_id\"] == '0n1w0w1VsRF4eqeqZHnX4g']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[1121, 1942, 4351, 7978, 5500]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "reviewer_rating = []\n",
    "ML_rating = []\n",
    "\n",
    "reviewer_rating.append(test_results_pandas_df[(test_results_pandas_df['label']==4.0)]['label'].count())\n",
    "ML_rating.append(test_results_pandas_df[(test_results_pandas_df['prediction']==4.0)]['prediction'].count())\n",
    "\n",
    "reviewer_rating.append(test_results_pandas_df[(test_results_pandas_df['label']==3.0)]['label'].count())\n",
    "ML_rating.append(test_results_pandas_df[(test_results_pandas_df['prediction']==3.0)]['prediction'].count())\n",
    "\n",
    "reviewer_rating.append(test_results_pandas_df[(test_results_pandas_df['label']==2.0)]['label'].count())\n",
    "ML_rating.append(test_results_pandas_df[(test_results_pandas_df['prediction']==2.0)]['prediction'].count())\n",
    "\n",
    "reviewer_rating.append(test_results_pandas_df[(test_results_pandas_df['label']==0.0)]['label'].count())\n",
    "ML_rating.append(test_results_pandas_df[(test_results_pandas_df['prediction']==0.0)]['prediction'].count())\n",
    "\n",
    "reviewer_rating.append(test_results_pandas_df[(test_results_pandas_df['label']==1.0)]['label'].count())\n",
    "ML_rating.append(test_results_pandas_df[(test_results_pandas_df['prediction']==1.0)]['prediction'].count())\n",
    " \n",
    "reviewer_rating    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[502, 1139, 3425, 11054, 4772]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ML_rating"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnoAAAIlCAYAAABCclUuAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1d5a43b39b0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(9, 8), dpi= 80, facecolor='w', edgecolor='k')\n",
    "\n",
    "x_range = np.arange(5)\n",
    "width = 0.25\n",
    "\n",
    "plt.bar(x_range+1, reviewer_rating, width)\n",
    "plt.bar(x_range+1 + width, ML_rating, width)\n",
    "\n",
    "\n",
    "plt.xlabel('Rating')\n",
    "plt.ylabel('Number of reviews')\n",
    "plt.title('Naive Bayes + Ngram VS reviewer rating')\n",
    "plt.legend(['NaiveBayes + Ngram Algurithm prediction of rating', 'The reviewer rating'])\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "test_results_pandas_df.to_csv(\"NavieBays_testresult.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "spark.stop()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}