sentimantal_analysis_on_movie_review/Kaggle_ Sentiment Analysis on Movie Reviews.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to /Users/song-\n",
      "[nltk_data]     yeongsug/nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.feature_extraction.text import CountVectorizer \n",
    "import re\n",
    "import nltk\n",
    "nltk.download('stopwords')\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.stem.porter import PorterStemmer\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import seaborn as sns\n",
    "from wordcloud import WordCloud,STOPWORDS\n",
    "from numpy import nan\n",
    "from bs4 import BeautifulSoup    \n",
    "from nltk.stem import WordNetLemmatizer\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from math import sqrt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "filepath = 'Users/song-yeongsug/workspace/kaggle_nlp/data_eng'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data = pd.read_csv('data_eng/train.tsv',sep='\\t')\n",
    "test_data = pd.read_csv('data_eng/train.tsv',sep='\\t')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 156060 entries, 0 to 156059\n",
      "Data columns (total 4 columns):\n",
      "PhraseId      156060 non-null int64\n",
      "SentenceId    156060 non-null int64\n",
      "Phrase        156060 non-null object\n",
      "Sentiment     156060 non-null int64\n",
      "dtypes: int64(3), object(1)\n",
      "memory usage: 4.8+ MB\n"
     ]
    }
   ],
   "source": [
    "train_data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 156060 entries, 0 to 156059\n",
      "Data columns (total 4 columns):\n",
      "PhraseId      156060 non-null int64\n",
      "SentenceId    156060 non-null int64\n",
      "Phrase        156060 non-null object\n",
      "Sentiment     156060 non-null int64\n",
      "dtypes: int64(3), object(1)\n",
      "memory usage: 4.8+ MB\n"
     ]
    }
   ],
   "source": [
    "test_data.info()"
   ]
  },
  {
   "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>PhraseId</th>\n",
       "      <th>SentenceId</th>\n",
       "      <th>Phrase</th>\n",
       "      <th>Sentiment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>A series of escapades demonstrating the adage ...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>A series of escapades demonstrating the adage ...</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>A series</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>A</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>series</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PhraseId  SentenceId                                             Phrase  \\\n",
       "0         1           1  A series of escapades demonstrating the adage ...   \n",
       "1         2           1  A series of escapades demonstrating the adage ...   \n",
       "2         3           1                                           A series   \n",
       "3         4           1                                                  A   \n",
       "4         5           1                                             series   \n",
       "\n",
       "   Sentiment  \n",
       "0          1  \n",
       "1          2  \n",
       "2          2  \n",
       "3          2  \n",
       "4          2  "
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "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>PhraseId</th>\n",
       "      <th>SentenceId</th>\n",
       "      <th>Phrase</th>\n",
       "      <th>Sentiment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>156055</th>\n",
       "      <td>156056</td>\n",
       "      <td>8544</td>\n",
       "      <td>Hearst 's</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>156056</th>\n",
       "      <td>156057</td>\n",
       "      <td>8544</td>\n",
       "      <td>forced avuncular chortles</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>156057</th>\n",
       "      <td>156058</td>\n",
       "      <td>8544</td>\n",
       "      <td>avuncular chortles</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>156058</th>\n",
       "      <td>156059</td>\n",
       "      <td>8544</td>\n",
       "      <td>avuncular</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>156059</th>\n",
       "      <td>156060</td>\n",
       "      <td>8544</td>\n",
       "      <td>chortles</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        PhraseId  SentenceId                     Phrase  Sentiment\n",
       "156055    156056        8544                  Hearst 's          2\n",
       "156056    156057        8544  forced avuncular chortles          1\n",
       "156057    156058        8544         avuncular chortles          3\n",
       "156058    156059        8544                  avuncular          2\n",
       "156059    156060        8544                   chortles          2"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# create new column with if and for loop with the Sentiment Phrase column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "Sentiment_words=[]\n",
    "for row in train_data['Sentiment']:\n",
    "    if row ==0:\n",
    "        Sentiment_words.append('negative')\n",
    "    elif row == 1:\n",
    "        Sentiment_words.append('neutral')\n",
    "    elif row == 2:\n",
    "        Sentiment_words.append('somewhat negative')\n",
    "    elif row == 3:\n",
    "        Sentiment_words.append('somewhat positive')\n",
    "    elif row == 4:\n",
    "        Sentiment_words.append('positive')\n",
    "    else:\n",
    "        Sentiment_words.append('Failed')\n",
    "train_data['Sentiment_words'] = Sentiment_words"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# count values of Sentiment Phrase"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "somewhat positive    32927\n",
       "positive              9206\n",
       "neutral              27273\n",
       "negative              7072\n",
       "somewhat negative    79582\n",
       "dtype: int64"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "word_count=pd.value_counts(train_data['Sentiment_words'].values, sort=False)\n",
    "word_count"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# some visually for Sentiment phrase"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1176a4860>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Index = [1,2,3,4,5]\n",
    "plt.figure(figsize=(15,5))\n",
    "plt.bar(Index,word_count,color = 'blue')\n",
    "plt.xticks(Index,['negative','neutral','somewhat negative','somewhat positive','positive'],rotation=45)\n",
    "plt.ylabel('word_count')\n",
    "plt.xlabel('word')\n",
    "plt.title('Count of Moods')\n",
    "plt.bar(Index, word_count)\n",
    "for a,b in zip(Index, word_count):\n",
    "    plt.text(a, b, str(b) ,color='black', fontweight='bold')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "# function to clean the column Phrase in the data set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "def review_to_words(raw_review): \n",
    "    review =raw_review\n",
    "    review = re.sub('[^a-zA-Z]', ' ',review)\n",
    "    review = review.lower()\n",
    "    review = review.split()\n",
    "    lemmatizer = WordNetLemmatizer()\n",
    "    review = [lemmatizer.lemmatize(w) for w in review if not w in set(stopwords.words('english'))]\n",
    "    return (' '.join(review))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "# run the function in train and test data set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "corpus= []\n",
    "for i in range(0, 156060):\n",
    "    corpus.append(review_to_words(train_data['Phrase'][i]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "corpus1= []\n",
    "for i in range(0, 156060):\n",
    "    corpus1.append(review_to_words(train_data['Phrase'][i]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "# create new column and merge it with the new cleaning list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data['new_phrase']=corpus"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Drop the old column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data.drop(['Phrase'],axis=1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "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>PhraseId</th>\n",
       "      <th>SentenceId</th>\n",
       "      <th>Sentiment</th>\n",
       "      <th>Sentiment_words</th>\n",
       "      <th>new_phrase</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>neutral</td>\n",
       "      <td>series escapade demonstrating adage good goose...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>somewhat negative</td>\n",
       "      <td>series escapade demonstrating adage good goose</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>somewhat negative</td>\n",
       "      <td>series</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>somewhat negative</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>somewhat negative</td>\n",
       "      <td>series</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   PhraseId  SentenceId  Sentiment    Sentiment_words  \\\n",
       "0         1           1          1            neutral   \n",
       "1         2           1          2  somewhat negative   \n",
       "2         3           1          2  somewhat negative   \n",
       "3         4           1          2  somewhat negative   \n",
       "4         5           1          2  somewhat negative   \n",
       "\n",
       "                                          new_phrase  \n",
       "0  series escapade demonstrating adage good goose...  \n",
       "1     series escapade demonstrating adage good goose  \n",
       "2                                             series  \n",
       "3                                                     \n",
       "4                                             series  "
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# select  positive Sentiment"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## select positive Sentiment from data set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "positive=train_data[train_data['Sentiment_words']==('positive')]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# prepare the data to split it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "words = ' '.join(positive['new_phrase'])\n",
    "split_word = \" \".join([word for word in words.split()])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# prepare the data to visual it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "wordcloud = WordCloud(stopwords=STOPWORDS,\n",
    "                      background_color='black',\n",
    "                      width=3000,\n",
    "                      height=2500\n",
    "                     ).generate(split_word)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x114b9def0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(1,figsize=(13, 13))\n",
    "plt.imshow(wordcloud)\n",
    "plt.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Now here some steps to count every word in postive Sentiment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "pos=positive['new_phrase']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "vectorizer = CountVectorizer(analyzer = \"word\",   \n",
    "                             tokenizer = None,    \n",
    "                             preprocessor = None, \n",
    "                             stop_words = None,   \n",
    "                             max_features = 2000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "pos_words = vectorizer.fit_transform(pos)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "pos_words = pos_words.toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "pos= vectorizer.get_feature_names()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['ability',\n",
       " 'able',\n",
       " 'absolute',\n",
       " 'absolutely',\n",
       " 'absorbing',\n",
       " 'accept',\n",
       " 'accessible',\n",
       " 'acclaim',\n",
       " 'accomplish',\n",
       " 'accomplished',\n",
       " 'accomplishes',\n",
       " 'accomplishment',\n",
       " 'accurate',\n",
       " 'achievement',\n",
       " 'achieves',\n",
       " 'achingly',\n",
       " 'across',\n",
       " 'act',\n",
       " 'acted',\n",
       " 'acting',\n",
       " 'action',\n",
       " 'actor',\n",
       " 'actress',\n",
       " 'actually',\n",
       " 'acute',\n",
       " 'adam',\n",
       " 'adaptation',\n",
       " 'add',\n",
       " 'addition',\n",
       " 'admirable',\n",
       " 'admire',\n",
       " 'adorably',\n",
       " 'adorns',\n",
       " 'adrenalin',\n",
       " 'adrenaline',\n",
       " 'adult',\n",
       " 'adventure',\n",
       " 'adventurous',\n",
       " 'aerial',\n",
       " 'affecting',\n",
       " 'affection',\n",
       " 'affirming',\n",
       " 'affirms',\n",
       " 'age',\n",
       " 'agenda',\n",
       " 'aggrandizing',\n",
       " 'ago',\n",
       " 'ahead',\n",
       " 'aimed',\n",
       " 'air',\n",
       " 'alike',\n",
       " 'alive',\n",
       " 'allegory',\n",
       " 'allen',\n",
       " 'allows',\n",
       " 'alluring',\n",
       " 'almost',\n",
       " 'alone',\n",
       " 'along',\n",
       " 'also',\n",
       " 'alternately',\n",
       " 'always',\n",
       " 'amazing',\n",
       " 'amazingly',\n",
       " 'ambiguity',\n",
       " 'ambitious',\n",
       " 'america',\n",
       " 'american',\n",
       " 'among',\n",
       " 'amount',\n",
       " 'amused',\n",
       " 'amusing',\n",
       " 'anchor',\n",
       " 'anderson',\n",
       " 'angst',\n",
       " 'animal',\n",
       " 'animated',\n",
       " 'animation',\n",
       " 'anime',\n",
       " 'another',\n",
       " 'answer',\n",
       " 'antic',\n",
       " 'anybody',\n",
       " 'anyone',\n",
       " 'anything',\n",
       " 'apart',\n",
       " 'apex',\n",
       " 'aplomb',\n",
       " 'apocalypse',\n",
       " 'apparent',\n",
       " 'appeal',\n",
       " 'appealing',\n",
       " 'appear',\n",
       " 'appreciate',\n",
       " 'approach',\n",
       " 'argument',\n",
       " 'around',\n",
       " 'art',\n",
       " 'artful',\n",
       " 'artfully',\n",
       " 'artifice',\n",
       " 'artist',\n",
       " 'artistic',\n",
       " 'artwork',\n",
       " 'ascends',\n",
       " 'asian',\n",
       " 'aspect',\n",
       " 'assembled',\n",
       " 'assurance',\n",
       " 'assured',\n",
       " 'astonish',\n",
       " 'astonishing',\n",
       " 'astonishingly',\n",
       " 'astounding',\n",
       " 'astoundingly',\n",
       " 'atlantic',\n",
       " 'atmosphere',\n",
       " 'attention',\n",
       " 'attitude',\n",
       " 'attract',\n",
       " 'attractive',\n",
       " 'audiard',\n",
       " 'audience',\n",
       " 'auteuil',\n",
       " 'authentic',\n",
       " 'average',\n",
       " 'avoid',\n",
       " 'awake',\n",
       " 'award',\n",
       " 'away',\n",
       " 'awe',\n",
       " 'awesome',\n",
       " 'back',\n",
       " 'backmasking',\n",
       " 'backstage',\n",
       " 'bad',\n",
       " 'balance',\n",
       " 'balanced',\n",
       " 'band',\n",
       " 'bang',\n",
       " 'baroque',\n",
       " 'base',\n",
       " 'based',\n",
       " 'battle',\n",
       " 'beach',\n",
       " 'beat',\n",
       " 'beautiful',\n",
       " 'beautifully',\n",
       " 'beauty',\n",
       " 'become',\n",
       " 'becomes',\n",
       " 'begin',\n",
       " 'behind',\n",
       " 'behold',\n",
       " 'believable',\n",
       " 'believe',\n",
       " 'belly',\n",
       " 'belongs',\n",
       " 'beloved',\n",
       " 'belt',\n",
       " 'beneath',\n",
       " 'best',\n",
       " 'better',\n",
       " 'beyond',\n",
       " 'bielinsky',\n",
       " 'big',\n",
       " 'bigger',\n",
       " 'biggest',\n",
       " 'bill',\n",
       " 'bio',\n",
       " 'bit',\n",
       " 'biting',\n",
       " 'bittersweet',\n",
       " 'black',\n",
       " 'blame',\n",
       " 'blend',\n",
       " 'bliss',\n",
       " 'blockbuster',\n",
       " 'blood',\n",
       " 'blue',\n",
       " 'body',\n",
       " 'bold',\n",
       " 'bond',\n",
       " 'bone',\n",
       " 'book',\n",
       " 'boring',\n",
       " 'born',\n",
       " 'bound',\n",
       " 'boundary',\n",
       " 'boy',\n",
       " 'bracing',\n",
       " 'brain',\n",
       " 'brainpower',\n",
       " 'brave',\n",
       " 'break',\n",
       " 'breakthrough',\n",
       " 'breath',\n",
       " 'breathtaking',\n",
       " 'breathtakingly',\n",
       " 'bride',\n",
       " 'bright',\n",
       " 'brilliant',\n",
       " 'brilliantly',\n",
       " 'bring',\n",
       " 'bringing',\n",
       " 'brings',\n",
       " 'brisk',\n",
       " 'british',\n",
       " 'brother',\n",
       " 'brought',\n",
       " 'brown',\n",
       " 'bruce',\n",
       " 'brush',\n",
       " 'brutal',\n",
       " 'brutally',\n",
       " 'bucked',\n",
       " 'budget',\n",
       " 'buff',\n",
       " 'build',\n",
       " 'bullock',\n",
       " 'buoyant',\n",
       " 'buried',\n",
       " 'burn',\n",
       " 'business',\n",
       " 'ca',\n",
       " 'caliber',\n",
       " 'call',\n",
       " 'called',\n",
       " 'calm',\n",
       " 'camera',\n",
       " 'candy',\n",
       " 'canvas',\n",
       " 'caper',\n",
       " 'captivated',\n",
       " 'captivating',\n",
       " 'capture',\n",
       " 'capturing',\n",
       " 'care',\n",
       " 'career',\n",
       " 'carmen',\n",
       " 'carol',\n",
       " 'carried',\n",
       " 'carry',\n",
       " 'case',\n",
       " 'cast',\n",
       " 'casting',\n",
       " 'cat',\n",
       " 'cautionary',\n",
       " 'celebrated',\n",
       " 'celebration',\n",
       " 'central',\n",
       " 'century',\n",
       " 'certain',\n",
       " 'certainly',\n",
       " 'certified',\n",
       " 'chabrol',\n",
       " 'challenging',\n",
       " 'champion',\n",
       " 'chance',\n",
       " 'change',\n",
       " 'changing',\n",
       " 'chaotic',\n",
       " 'character',\n",
       " 'characterization',\n",
       " 'charged',\n",
       " 'charisma',\n",
       " 'charismatic',\n",
       " 'charm',\n",
       " 'charmer',\n",
       " 'charming',\n",
       " 'chase',\n",
       " 'cheap',\n",
       " 'cheeky',\n",
       " 'cheer',\n",
       " 'chemistry',\n",
       " 'chick',\n",
       " 'chief',\n",
       " 'child',\n",
       " 'childhood',\n",
       " 'chilling',\n",
       " 'china',\n",
       " 'chinese',\n",
       " 'choice',\n",
       " 'chou',\n",
       " 'chris',\n",
       " 'christian',\n",
       " 'christmas',\n",
       " 'christopher',\n",
       " 'cinema',\n",
       " 'cinematic',\n",
       " 'cinematography',\n",
       " 'city',\n",
       " 'clarity',\n",
       " 'clashing',\n",
       " 'class',\n",
       " 'classic',\n",
       " 'clear',\n",
       " 'clearly',\n",
       " 'clever',\n",
       " 'cleverest',\n",
       " 'cleverly',\n",
       " 'cliche',\n",
       " 'clinic',\n",
       " 'co',\n",
       " 'coheres',\n",
       " 'cohesive',\n",
       " 'cold',\n",
       " 'collectively',\n",
       " 'color',\n",
       " 'colorful',\n",
       " 'colour',\n",
       " 'combination',\n",
       " 'combine',\n",
       " 'come',\n",
       " 'comedic',\n",
       " 'comedy',\n",
       " 'comic',\n",
       " 'coming',\n",
       " 'communal',\n",
       " 'company',\n",
       " 'compassion',\n",
       " 'compelling',\n",
       " 'compellingly',\n",
       " 'completely',\n",
       " 'complex',\n",
       " 'complexity',\n",
       " 'complexly',\n",
       " 'complicated',\n",
       " 'computer',\n",
       " 'con',\n",
       " 'conceit',\n",
       " 'concept',\n",
       " 'concert',\n",
       " 'conclusion',\n",
       " 'confident',\n",
       " 'conflict',\n",
       " 'confusing',\n",
       " 'consider',\n",
       " 'considerable',\n",
       " 'considered',\n",
       " 'consistently',\n",
       " 'constant',\n",
       " 'constantly',\n",
       " 'construct',\n",
       " 'constructed',\n",
       " 'contemporary',\n",
       " 'contribution',\n",
       " 'control',\n",
       " 'convey',\n",
       " 'conveying',\n",
       " 'conviction',\n",
       " 'convincing',\n",
       " 'cool',\n",
       " 'core',\n",
       " 'costume',\n",
       " 'could',\n",
       " 'country',\n",
       " 'couple',\n",
       " 'coupled',\n",
       " 'courage',\n",
       " 'course',\n",
       " 'cover',\n",
       " 'crack',\n",
       " 'crafted',\n",
       " 'crafty',\n",
       " 'cranky',\n",
       " 'create',\n",
       " 'created',\n",
       " 'creates',\n",
       " 'creating',\n",
       " 'creation',\n",
       " 'creative',\n",
       " 'creativity',\n",
       " 'credit',\n",
       " 'creepy',\n",
       " 'crime',\n",
       " 'crisp',\n",
       " 'crowd',\n",
       " 'crucial',\n",
       " 'crush',\n",
       " 'cuisine',\n",
       " 'cult',\n",
       " 'cultural',\n",
       " 'culture',\n",
       " 'cut',\n",
       " 'cutting',\n",
       " 'daily',\n",
       " 'damn',\n",
       " 'damned',\n",
       " 'dancing',\n",
       " 'daniel',\n",
       " 'dare',\n",
       " 'daring',\n",
       " 'dark',\n",
       " 'darkly',\n",
       " 'date',\n",
       " 'daughter',\n",
       " 'david',\n",
       " 'day',\n",
       " 'dazzle',\n",
       " 'dazzling',\n",
       " 'de',\n",
       " 'dead',\n",
       " 'deadpan',\n",
       " 'deal',\n",
       " 'dealing',\n",
       " 'death',\n",
       " 'debut',\n",
       " 'decade',\n",
       " 'decent',\n",
       " 'deceptively',\n",
       " 'deep',\n",
       " 'deeper',\n",
       " 'deeply',\n",
       " 'definitely',\n",
       " 'deft',\n",
       " 'deftly',\n",
       " 'delectable',\n",
       " 'delicate',\n",
       " 'delicately',\n",
       " 'delicious',\n",
       " 'delight',\n",
       " 'delightful',\n",
       " 'delightfully',\n",
       " 'deliver',\n",
       " 'delivered',\n",
       " 'delivering',\n",
       " 'delivers',\n",
       " 'delivery',\n",
       " 'demand',\n",
       " 'demon',\n",
       " 'demonstrates',\n",
       " 'depiction',\n",
       " 'depth',\n",
       " 'derivative',\n",
       " 'deserved',\n",
       " 'deserves',\n",
       " 'deserving',\n",
       " 'design',\n",
       " 'despite',\n",
       " 'destination',\n",
       " 'destined',\n",
       " 'detail',\n",
       " 'detailed',\n",
       " 'devastating',\n",
       " 'device',\n",
       " 'dialogue',\n",
       " 'diane',\n",
       " 'dickens',\n",
       " 'die',\n",
       " 'diesel',\n",
       " 'different',\n",
       " 'difficult',\n",
       " 'dignity',\n",
       " 'dimensional',\n",
       " 'directed',\n",
       " 'directing',\n",
       " 'direction',\n",
       " 'director',\n",
       " 'directorial',\n",
       " 'directs',\n",
       " 'discover',\n",
       " 'discovery',\n",
       " 'disney',\n",
       " 'display',\n",
       " 'distinctive',\n",
       " 'distinguished',\n",
       " 'diverse',\n",
       " 'divine',\n",
       " 'document',\n",
       " 'documentary',\n",
       " 'dog',\n",
       " 'done',\n",
       " 'doubt',\n",
       " 'downright',\n",
       " 'dragon',\n",
       " 'drama',\n",
       " 'dramatic',\n",
       " 'dramatically',\n",
       " 'draw',\n",
       " 'drawing',\n",
       " 'drawn',\n",
       " 'dream',\n",
       " 'drenched',\n",
       " 'driven',\n",
       " 'droll',\n",
       " 'dropping',\n",
       " 'dry',\n",
       " 'dull',\n",
       " 'dumb',\n",
       " 'dustin',\n",
       " 'duty',\n",
       " 'dying',\n",
       " 'early',\n",
       " 'earned',\n",
       " 'earnest',\n",
       " 'earth',\n",
       " 'easily',\n",
       " 'easy',\n",
       " 'edge',\n",
       " 'edited',\n",
       " 'eerie',\n",
       " 'effect',\n",
       " 'effective',\n",
       " 'effectively',\n",
       " 'effort',\n",
       " 'effortless',\n",
       " 'effortlessly',\n",
       " 'either',\n",
       " 'elaborate',\n",
       " 'elegance',\n",
       " 'elegant',\n",
       " 'elegantly',\n",
       " 'element',\n",
       " 'ellen',\n",
       " 'eloquent',\n",
       " 'else',\n",
       " 'embrace',\n",
       " 'embracing',\n",
       " 'emerge',\n",
       " 'emerges',\n",
       " 'emotion',\n",
       " 'emotional',\n",
       " 'emotionally',\n",
       " 'empire',\n",
       " 'enchanting',\n",
       " 'encounter',\n",
       " 'encourage',\n",
       " 'encouraging',\n",
       " 'end',\n",
       " 'endearing',\n",
       " 'ending',\n",
       " 'enduring',\n",
       " 'energetic',\n",
       " 'energy',\n",
       " 'engaging',\n",
       " 'engrossing',\n",
       " 'enhances',\n",
       " 'enjoy',\n",
       " 'enjoyable',\n",
       " 'enjoyably',\n",
       " 'enjoyed',\n",
       " 'enlightening',\n",
       " 'enormous',\n",
       " 'enormously',\n",
       " 'enough',\n",
       " 'enriched',\n",
       " 'ensemble',\n",
       " 'enter',\n",
       " 'enterprise',\n",
       " 'entertain',\n",
       " 'entertained',\n",
       " 'entertaining',\n",
       " 'entertainment',\n",
       " 'enthralling',\n",
       " 'enthusiasm',\n",
       " 'enthusiastic',\n",
       " 'entire',\n",
       " 'entirely',\n",
       " 'entry',\n",
       " 'enveloping',\n",
       " 'environmental',\n",
       " 'epic',\n",
       " 'equally',\n",
       " 'era',\n",
       " 'especially',\n",
       " 'esque',\n",
       " 'ethereal',\n",
       " 'even',\n",
       " 'evening',\n",
       " 'event',\n",
       " 'ever',\n",
       " 'every',\n",
       " 'everyone',\n",
       " 'everything',\n",
       " 'evil',\n",
       " 'evocative',\n",
       " 'examination',\n",
       " 'examines',\n",
       " 'example',\n",
       " 'exceedingly',\n",
       " 'excellent',\n",
       " 'excels',\n",
       " 'exceptional',\n",
       " 'exceptionally',\n",
       " 'excitement',\n",
       " 'excites',\n",
       " 'exciting',\n",
       " 'executed',\n",
       " 'execution',\n",
       " 'exercise',\n",
       " 'exhilarating',\n",
       " 'existence',\n",
       " 'exotic',\n",
       " 'expect',\n",
       " 'expectation',\n",
       " 'experience',\n",
       " 'expert',\n",
       " 'expertly',\n",
       " 'exploitative',\n",
       " 'exploration',\n",
       " 'explores',\n",
       " 'express',\n",
       " 'expressive',\n",
       " 'exquisite',\n",
       " 'exterior',\n",
       " 'extraordinarily',\n",
       " 'extraordinary',\n",
       " 'extreme',\n",
       " 'extremely',\n",
       " 'eye',\n",
       " 'eyed',\n",
       " 'fabulous',\n",
       " 'face',\n",
       " 'fact',\n",
       " 'fairly',\n",
       " 'fairy',\n",
       " 'faith',\n",
       " 'faithful',\n",
       " 'fall',\n",
       " 'familiar',\n",
       " 'family',\n",
       " 'fan',\n",
       " 'fanciful',\n",
       " 'fantastic',\n",
       " 'fantasy',\n",
       " 'far',\n",
       " 'fare',\n",
       " 'fascinating',\n",
       " 'fashion',\n",
       " 'fashioned',\n",
       " 'fast',\n",
       " 'fat',\n",
       " 'father',\n",
       " 'favorite',\n",
       " 'fear',\n",
       " 'fearlessness',\n",
       " 'feast',\n",
       " 'feat',\n",
       " 'feature',\n",
       " 'featuring',\n",
       " 'feel',\n",
       " 'feeling',\n",
       " 'felt',\n",
       " 'female',\n",
       " 'ferrara',\n",
       " 'festival',\n",
       " 'fi',\n",
       " 'fiercely',\n",
       " 'fight',\n",
       " 'filled',\n",
       " 'filling',\n",
       " 'film',\n",
       " 'filmed',\n",
       " 'filming',\n",
       " 'filmmaker',\n",
       " 'filmmaking',\n",
       " 'finally',\n",
       " 'find',\n",
       " 'fine',\n",
       " 'finely',\n",
       " 'finest',\n",
       " 'finish',\n",
       " 'fire',\n",
       " 'first',\n",
       " 'fish',\n",
       " 'fit',\n",
       " 'flair',\n",
       " 'flash',\n",
       " 'flat',\n",
       " 'flavor',\n",
       " 'flawed',\n",
       " 'flawless',\n",
       " 'flick',\n",
       " 'flow',\n",
       " 'fluid',\n",
       " 'folk',\n",
       " 'follow',\n",
       " 'fondness',\n",
       " 'fontaine',\n",
       " 'food',\n",
       " 'footage',\n",
       " 'force',\n",
       " 'forceful',\n",
       " 'forcefully',\n",
       " 'foreign',\n",
       " 'forget',\n",
       " 'form',\n",
       " 'format',\n",
       " 'formed',\n",
       " 'formula',\n",
       " 'found',\n",
       " 'four',\n",
       " 'fourth',\n",
       " 'frailty',\n",
       " 'frame',\n",
       " 'franchise',\n",
       " 'frank',\n",
       " 'free',\n",
       " 'french',\n",
       " 'frenetic',\n",
       " 'frequent',\n",
       " 'frequently',\n",
       " 'fresh',\n",
       " 'freshness',\n",
       " 'friendship',\n",
       " 'front',\n",
       " 'frozen',\n",
       " 'fulfill',\n",
       " 'fulfilling',\n",
       " 'full',\n",
       " 'fully',\n",
       " 'fun',\n",
       " 'funnier',\n",
       " 'funniest',\n",
       " 'funny',\n",
       " 'fused',\n",
       " 'future',\n",
       " 'fuzzy',\n",
       " 'gag',\n",
       " 'gai',\n",
       " 'game',\n",
       " 'gang',\n",
       " 'gangster',\n",
       " 'gasp',\n",
       " 'gasping',\n",
       " 'gay',\n",
       " 'gem',\n",
       " 'gender',\n",
       " 'general',\n",
       " 'generosity',\n",
       " 'generous',\n",
       " 'genius',\n",
       " 'genre',\n",
       " 'gentle',\n",
       " 'gently',\n",
       " 'genuine',\n",
       " 'genuinely',\n",
       " 'gere',\n",
       " 'get',\n",
       " 'getting',\n",
       " 'ghetto',\n",
       " 'giddy',\n",
       " 'gift',\n",
       " 'gifted',\n",
       " 'girl',\n",
       " 'give',\n",
       " 'given',\n",
       " 'giving',\n",
       " 'glimpse',\n",
       " 'glorious',\n",
       " 'gloriously',\n",
       " 'glued',\n",
       " 'go',\n",
       " 'goal',\n",
       " 'god',\n",
       " 'godard',\n",
       " 'goer',\n",
       " 'going',\n",
       " 'goo',\n",
       " 'good',\n",
       " 'goodfellas',\n",
       " 'gorgeous',\n",
       " 'gorgeously',\n",
       " 'gosling',\n",
       " 'got',\n",
       " 'grace',\n",
       " 'graceful',\n",
       " 'gracefully',\n",
       " 'gradually',\n",
       " 'grand',\n",
       " 'grandeur',\n",
       " 'grant',\n",
       " 'graphic',\n",
       " 'great',\n",
       " 'greater',\n",
       " 'greatest',\n",
       " 'greatly',\n",
       " 'greatness',\n",
       " 'greek',\n",
       " 'grew',\n",
       " 'grief',\n",
       " 'griffith',\n",
       " 'grip',\n",
       " 'gripping',\n",
       " 'gritty',\n",
       " 'group',\n",
       " 'grow',\n",
       " 'growing',\n",
       " 'grown',\n",
       " 'guaranteed',\n",
       " 'guessing',\n",
       " 'guilt',\n",
       " 'guilty',\n",
       " 'gut',\n",
       " 'guy',\n",
       " 'gyllenhaal',\n",
       " 'hair',\n",
       " 'half',\n",
       " 'hand',\n",
       " 'handed',\n",
       " 'handled',\n",
       " 'handsome',\n",
       " 'happiness',\n",
       " 'happy',\n",
       " 'hard',\n",
       " 'haunting',\n",
       " 'head',\n",
       " 'headed',\n",
       " 'heady',\n",
       " 'heard',\n",
       " 'heart',\n",
       " 'heartbreaking',\n",
       " 'hearted',\n",
       " 'heartfelt',\n",
       " 'heartwarming',\n",
       " 'heated',\n",
       " 'heaven',\n",
       " 'heightened',\n",
       " 'held',\n",
       " 'help',\n",
       " 'hennings',\n",
       " 'herald',\n",
       " 'hero',\n",
       " 'herzog',\n",
       " 'high',\n",
       " 'highest',\n",
       " 'highly',\n",
       " 'hilarious',\n",
       " 'hilariously',\n",
       " 'historical',\n",
       " 'historically',\n",
       " 'history',\n",
       " 'hit',\n",
       " 'hitting',\n",
       " 'hoffman',\n",
       " 'hold',\n",
       " 'holiday',\n",
       " 'holly',\n",
       " 'hollywood',\n",
       " 'holm',\n",
       " 'holocaust',\n",
       " 'holographic',\n",
       " 'home',\n",
       " 'honest',\n",
       " 'honesty',\n",
       " 'hong',\n",
       " 'honored',\n",
       " 'hoot',\n",
       " 'hope',\n",
       " 'hopeful',\n",
       " 'hopefully',\n",
       " 'hoping',\n",
       " 'horrifying',\n",
       " 'horror',\n",
       " 'host',\n",
       " 'hot',\n",
       " 'hour',\n",
       " 'house',\n",
       " 'however',\n",
       " 'huge',\n",
       " 'hugely',\n",
       " 'human',\n",
       " 'humanity',\n",
       " 'humor',\n",
       " 'humorous',\n",
       " 'humour',\n",
       " 'hunter',\n",
       " 'huppert',\n",
       " 'hypnotic',\n",
       " 'ian',\n",
       " 'ice',\n",
       " 'idea',\n",
       " 'idealistic',\n",
       " 'identity',\n",
       " 'ii',\n",
       " 'illuminating',\n",
       " 'image',\n",
       " 'imagery',\n",
       " 'imagination',\n",
       " 'imaginative',\n",
       " 'imaginatively',\n",
       " 'imagine',\n",
       " 'imax',\n",
       " 'immensely',\n",
       " 'impact',\n",
       " 'impeccable',\n",
       " 'important',\n",
       " 'impossible',\n",
       " 'impressed',\n",
       " 'impression',\n",
       " 'impressive',\n",
       " 'improbable',\n",
       " 'improves',\n",
       " 'improvised',\n",
       " 'impulse',\n",
       " 'including',\n",
       " 'incredible',\n",
       " 'incredibly',\n",
       " 'indian',\n",
       " 'indie',\n",
       " 'inducing',\n",
       " 'industry',\n",
       " 'inescapably',\n",
       " 'infectious',\n",
       " 'informative',\n",
       " 'ingenious',\n",
       " 'inimitable',\n",
       " 'injects',\n",
       " 'innocence',\n",
       " 'insane',\n",
       " 'insanely',\n",
       " 'inside',\n",
       " 'insight',\n",
       " 'insightful',\n",
       " 'insightfully',\n",
       " 'inspirational',\n",
       " 'inspired',\n",
       " 'inspiring',\n",
       " 'instantly',\n",
       " 'instead',\n",
       " 'instinct',\n",
       " 'integrity',\n",
       " 'intelligence',\n",
       " 'intelligent',\n",
       " 'intelligently',\n",
       " 'intended',\n",
       " 'intense',\n",
       " 'intensity',\n",
       " 'interest',\n",
       " 'interesting',\n",
       " 'interestingly',\n",
       " 'international',\n",
       " 'interview',\n",
       " 'intimacy',\n",
       " 'intimate',\n",
       " 'intoxicating',\n",
       " 'intricate',\n",
       " 'intrigue',\n",
       " 'intriguing',\n",
       " 'introduce',\n",
       " 'introduction',\n",
       " 'inuit',\n",
       " 'invaluable',\n",
       " 'inventive',\n",
       " 'inventiveness',\n",
       " 'inviting',\n",
       " 'involved',\n",
       " 'involving',\n",
       " 'irresistible',\n",
       " 'jackson',\n",
       " 'jacques',\n",
       " 'jake',\n",
       " 'james',\n",
       " 'japanese',\n",
       " 'jaw',\n",
       " 'jazzy',\n",
       " 'jealousy',\n",
       " 'jessica',\n",
       " 'job',\n",
       " 'john',\n",
       " 'joke',\n",
       " 'jolt',\n",
       " 'jones',\n",
       " 'journey',\n",
       " 'joy',\n",
       " 'joyous',\n",
       " 'jumble',\n",
       " 'jump',\n",
       " 'junior',\n",
       " 'justice',\n",
       " 'kaufman',\n",
       " 'keep',\n",
       " 'key',\n",
       " 'kid',\n",
       " 'kidman',\n",
       " 'killer',\n",
       " 'kind',\n",
       " 'kindness',\n",
       " 'kinetic',\n",
       " 'kinetically',\n",
       " 'king',\n",
       " 'kissing',\n",
       " 'kline',\n",
       " 'know',\n",
       " 'knowledge',\n",
       " 'known',\n",
       " 'kong',\n",
       " 'korean',\n",
       " 'krawczyk',\n",
       " 'kwan',\n",
       " 'lacking',\n",
       " 'landscape',\n",
       " 'lane',\n",
       " 'large',\n",
       " 'larger',\n",
       " 'last',\n",
       " 'laugh',\n",
       " 'laughed',\n",
       " 'laughing',\n",
       " ...]"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(9206, 2000)\n"
     ]
    }
   ],
   "source": [
    "print (pos_words.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ability 20\n",
      "able 32\n",
      "absolute 9\n",
      "absolutely 24\n",
      "absorbing 18\n",
      "accept 10\n",
      "accessible 16\n",
      "acclaim 9\n",
      "accomplish 5\n",
      "accomplished 23\n",
      "accomplishes 7\n",
      "accomplishment 17\n",
      "accurate 9\n",
      "achievement 30\n",
      "achieves 5\n",
      "achingly 7\n",
      "across 12\n",
      "act 29\n",
      "acted 43\n",
      "acting 101\n",
      "action 93\n",
      "actor 130\n",
      "actress 14\n",
      "actually 45\n",
      "acute 8\n",
      "adam 7\n",
      "adaptation 22\n",
      "add 17\n",
      "addition 13\n",
      "admirable 16\n",
      "admire 7\n",
      "adorably 7\n",
      "adorns 7\n",
      "adrenalin 6\n",
      "adrenaline 6\n",
      "adult 35\n",
      "adventure 54\n",
      "adventurous 17\n",
      "aerial 7\n",
      "affecting 9\n",
      "affection 11\n",
      "affirming 10\n",
      "affirms 5\n",
      "age 84\n",
      "agenda 7\n",
      "aggrandizing 7\n",
      "ago 9\n",
      "ahead 10\n",
      "aimed 5\n",
      "air 10\n",
      "alike 5\n",
      "alive 20\n",
      "allegory 7\n",
      "allen 15\n",
      "allows 19\n",
      "alluring 7\n",
      "almost 50\n",
      "alone 8\n",
      "along 29\n",
      "also 88\n",
      "alternately 6\n",
      "always 43\n",
      "amazing 35\n",
      "amazingly 11\n",
      "ambiguity 6\n",
      "ambitious 18\n",
      "america 18\n",
      "american 63\n",
      "among 19\n",
      "amount 16\n",
      "amused 9\n",
      "amusing 57\n",
      "anchor 6\n",
      "anderson 6\n",
      "angst 10\n",
      "animal 10\n",
      "animated 28\n",
      "animation 24\n",
      "anime 6\n",
      "another 43\n",
      "answer 11\n",
      "antic 7\n",
      "anybody 5\n",
      "anyone 35\n",
      "anything 31\n",
      "apart 8\n",
      "apex 5\n",
      "aplomb 11\n",
      "apocalypse 8\n",
      "apparent 6\n",
      "appeal 15\n",
      "appealing 27\n",
      "appear 8\n",
      "appreciate 12\n",
      "approach 9\n",
      "argument 6\n",
      "around 24\n",
      "art 59\n",
      "artful 8\n",
      "artfully 6\n",
      "artifice 8\n",
      "artist 41\n",
      "artistic 10\n",
      "artwork 7\n",
      "ascends 7\n",
      "asian 7\n",
      "aspect 10\n",
      "assembled 9\n",
      "assurance 9\n",
      "assured 17\n",
      "astonish 7\n",
      "astonishing 18\n",
      "astonishingly 11\n",
      "astounding 18\n",
      "astoundingly 8\n",
      "atlantic 6\n",
      "atmosphere 9\n",
      "attention 11\n",
      "attitude 6\n",
      "attract 6\n",
      "attractive 12\n",
      "audiard 7\n",
      "audience 97\n",
      "auteuil 8\n",
      "authentic 7\n",
      "average 7\n",
      "avoid 7\n",
      "awake 6\n",
      "award 7\n",
      "away 64\n",
      "awe 20\n",
      "awesome 7\n",
      "back 33\n",
      "backmasking 11\n",
      "backstage 8\n",
      "bad 7\n",
      "balance 15\n",
      "balanced 8\n",
      "band 10\n",
      "bang 6\n",
      "baroque 6\n",
      "base 6\n",
      "based 10\n",
      "battle 8\n",
      "beach 7\n",
      "beat 11\n",
      "beautiful 112\n",
      "beautifully 111\n",
      "beauty 42\n",
      "become 27\n",
      "becomes 9\n",
      "begin 7\n",
      "behind 7\n",
      "behold 14\n",
      "believable 24\n",
      "believe 12\n",
      "belly 10\n",
      "belongs 11\n",
      "beloved 6\n",
      "belt 6\n",
      "beneath 11\n",
      "best 394\n",
      "better 61\n",
      "beyond 18\n",
      "bielinsky 6\n",
      "big 76\n",
      "bigger 8\n",
      "biggest 6\n",
      "bill 10\n",
      "bio 9\n",
      "bit 41\n",
      "biting 6\n",
      "bittersweet 9\n",
      "black 27\n",
      "blame 9\n",
      "blend 10\n",
      "bliss 8\n",
      "blockbuster 11\n",
      "blood 6\n",
      "blue 21\n",
      "body 8\n",
      "bold 7\n",
      "bond 22\n",
      "bone 11\n",
      "book 12\n",
      "boring 6\n",
      "born 13\n",
      "bound 7\n",
      "boundary 6\n",
      "boy 16\n",
      "bracing 11\n",
      "brain 6\n",
      "brainpower 8\n",
      "brave 6\n",
      "break 7\n",
      "breakthrough 14\n",
      "breath 6\n",
      "breathtaking 29\n",
      "breathtakingly 28\n",
      "bride 6\n",
      "bright 16\n",
      "brilliant 94\n",
      "brilliantly 16\n",
      "bring 21\n",
      "bringing 12\n",
      "brings 23\n",
      "brisk 7\n",
      "british 14\n",
      "brother 17\n",
      "brought 9\n",
      "brown 16\n",
      "bruce 7\n",
      "brush 12\n",
      "brutal 13\n",
      "brutally 6\n",
      "bucked 8\n",
      "budget 8\n",
      "buff 7\n",
      "build 9\n",
      "bullock 14\n",
      "buoyant 13\n",
      "buried 6\n",
      "burn 11\n",
      "business 7\n",
      "ca 35\n",
      "caliber 7\n",
      "call 14\n",
      "called 12\n",
      "calm 6\n",
      "camera 27\n",
      "candy 7\n",
      "canvas 7\n",
      "caper 6\n",
      "captivated 6\n",
      "captivating 27\n",
      "capture 43\n",
      "capturing 9\n",
      "care 26\n",
      "career 43\n",
      "carmen 5\n",
      "carol 5\n",
      "carried 9\n",
      "carry 15\n",
      "case 8\n",
      "cast 146\n",
      "casting 22\n",
      "cat 10\n",
      "cautionary 6\n",
      "celebrated 6\n",
      "celebration 13\n",
      "central 6\n",
      "century 10\n",
      "certain 7\n",
      "certainly 16\n",
      "certified 5\n",
      "chabrol 7\n",
      "challenging 11\n",
      "champion 5\n",
      "chance 8\n",
      "change 6\n",
      "changing 11\n",
      "chaotic 5\n",
      "character 215\n",
      "characterization 8\n",
      "charged 7\n",
      "charisma 5\n",
      "charismatic 14\n",
      "charm 44\n",
      "charmer 11\n",
      "charming 94\n",
      "chase 9\n",
      "cheap 8\n",
      "cheeky 8\n",
      "cheer 8\n",
      "chemistry 17\n",
      "chick 6\n",
      "chief 5\n",
      "child 29\n",
      "childhood 13\n",
      "chilling 16\n",
      "china 7\n",
      "chinese 7\n",
      "choice 9\n",
      "chou 6\n",
      "chris 6\n",
      "christian 5\n",
      "christmas 7\n",
      "christopher 7\n",
      "cinema 95\n",
      "cinematic 65\n",
      "cinematography 38\n",
      "city 6\n",
      "clarity 16\n",
      "clashing 6\n",
      "class 35\n",
      "classic 44\n",
      "clear 12\n",
      "clearly 11\n",
      "clever 67\n",
      "cleverest 7\n",
      "cleverly 20\n",
      "cliche 7\n",
      "clinic 6\n",
      "co 17\n",
      "coheres 7\n",
      "cohesive 10\n",
      "cold 6\n",
      "collectively 9\n",
      "color 26\n",
      "colorful 32\n",
      "colour 7\n",
      "combination 9\n",
      "combine 14\n",
      "come 76\n",
      "comedic 10\n",
      "comedy 271\n",
      "comic 54\n",
      "coming 39\n",
      "communal 13\n",
      "company 10\n",
      "compassion 9\n",
      "compelling 90\n",
      "compellingly 7\n",
      "completely 24\n",
      "complex 24\n",
      "complexity 7\n",
      "complexly 8\n",
      "complicated 20\n",
      "computer 13\n",
      "con 6\n",
      "conceit 6\n",
      "concept 17\n",
      "concert 6\n",
      "conclusion 6\n",
      "confident 16\n",
      "conflict 7\n",
      "confusing 6\n",
      "consider 6\n",
      "considerable 26\n",
      "considered 7\n",
      "consistently 22\n",
      "constant 14\n",
      "constantly 7\n",
      "construct 6\n",
      "constructed 17\n",
      "contemporary 21\n",
      "contribution 7\n",
      "control 7\n",
      "convey 6\n",
      "conveying 6\n",
      "conviction 16\n",
      "convincing 28\n",
      "cool 34\n",
      "core 11\n",
      "costume 8\n",
      "could 54\n",
      "country 10\n",
      "couple 14\n",
      "coupled 7\n",
      "courage 15\n",
      "course 6\n",
      "cover 6\n",
      "crack 6\n",
      "crafted 51\n",
      "crafty 6\n",
      "cranky 7\n",
      "create 21\n",
      "created 21\n",
      "creates 22\n",
      "creating 6\n",
      "creation 9\n",
      "creative 44\n",
      "creativity 16\n",
      "credit 32\n",
      "creepy 17\n",
      "crime 31\n",
      "crisp 14\n",
      "crowd 22\n",
      "crucial 7\n",
      "crush 14\n",
      "cuisine 6\n",
      "cult 11\n",
      "cultural 9\n",
      "culture 40\n",
      "cut 14\n",
      "cutting 6\n",
      "daily 7\n",
      "damn 11\n",
      "damned 11\n",
      "dancing 7\n",
      "daniel 10\n",
      "dare 11\n",
      "daring 17\n",
      "dark 26\n",
      "darkly 6\n",
      "date 26\n",
      "daughter 10\n",
      "david 19\n",
      "day 33\n",
      "dazzle 9\n",
      "dazzling 38\n",
      "de 22\n",
      "dead 11\n",
      "deadpan 6\n",
      "deal 12\n",
      "dealing 6\n",
      "death 8\n",
      "debut 38\n",
      "decade 17\n",
      "decent 7\n",
      "deceptively 16\n",
      "deep 15\n",
      "deeper 8\n",
      "deeply 60\n",
      "definitely 19\n",
      "deft 16\n",
      "deftly 21\n",
      "delectable 6\n",
      "delicate 10\n",
      "delicately 10\n",
      "delicious 17\n",
      "delight 58\n",
      "delightful 56\n",
      "delightfully 26\n",
      "deliver 17\n",
      "delivered 9\n",
      "delivering 16\n",
      "delivers 30\n",
      "delivery 10\n",
      "demand 9\n",
      "demon 8\n",
      "demonstrates 8\n",
      "depiction 9\n",
      "depth 21\n",
      "derivative 7\n",
      "deserved 14\n",
      "deserves 28\n",
      "deserving 6\n",
      "design 7\n",
      "despite 12\n",
      "destination 9\n",
      "destined 7\n",
      "detail 16\n",
      "detailed 28\n",
      "devastating 14\n",
      "device 11\n",
      "dialogue 48\n",
      "diane 8\n",
      "dickens 11\n",
      "die 15\n",
      "diesel 6\n",
      "different 36\n",
      "difficult 7\n",
      "dignity 8\n",
      "dimensional 6\n",
      "directed 42\n",
      "directing 6\n",
      "direction 52\n",
      "director 143\n",
      "directorial 18\n",
      "directs 8\n",
      "discover 7\n",
      "discovery 13\n",
      "disney 26\n",
      "display 10\n",
      "distinctive 6\n",
      "distinguished 7\n",
      "diverse 6\n",
      "divine 6\n",
      "document 12\n",
      "documentary 117\n",
      "dog 8\n",
      "done 51\n",
      "doubt 11\n",
      "downright 6\n",
      "dragon 8\n",
      "drama 147\n",
      "dramatic 28\n",
      "dramatically 21\n",
      "draw 10\n",
      "drawing 8\n",
      "drawn 14\n",
      "dream 21\n",
      "drenched 7\n",
      "driven 10\n",
      "droll 7\n",
      "dropping 14\n",
      "dry 12\n",
      "dull 6\n",
      "dumb 10\n",
      "dustin 10\n",
      "duty 6\n",
      "dying 10\n",
      "early 6\n",
      "earned 6\n",
      "earnest 9\n",
      "earth 8\n",
      "easily 31\n",
      "easy 12\n",
      "edge 41\n",
      "edited 16\n",
      "eerie 9\n",
      "effect 35\n",
      "effective 10\n",
      "effectively 6\n",
      "effort 22\n",
      "effortless 8\n",
      "effortlessly 12\n",
      "either 12\n",
      "elaborate 9\n",
      "elegance 11\n",
      "elegant 26\n",
      "elegantly 9\n",
      "element 14\n",
      "ellen 9\n",
      "eloquent 15\n",
      "else 10\n",
      "embrace 9\n",
      "embracing 7\n",
      "emerge 14\n",
      "emerges 13\n",
      "emotion 27\n",
      "emotional 72\n",
      "emotionally 40\n",
      "empire 6\n",
      "enchanting 6\n",
      "encounter 7\n",
      "encourage 6\n",
      "encouraging 6\n",
      "end 49\n",
      "endearing 23\n",
      "ending 17\n",
      "enduring 8\n",
      "energetic 30\n",
      "energy 44\n",
      "engaging 94\n",
      "engrossing 49\n",
      "enhances 6\n",
      "enjoy 27\n",
      "enjoyable 111\n",
      "enjoyably 6\n",
      "enjoyed 17\n",
      "enlightening 9\n",
      "enormous 8\n",
      "enormously 15\n",
      "enough 82\n",
      "enriched 9\n",
      "ensemble 39\n",
      "enter 10\n",
      "enterprise 6\n",
      "entertain 16\n",
      "entertained 7\n",
      "entertaining 168\n",
      "entertainment 72\n",
      "enthralling 13\n",
      "enthusiasm 15\n",
      "enthusiastic 15\n",
      "entire 9\n",
      "entirely 16\n",
      "entry 7\n",
      "enveloping 10\n",
      "environmental 12\n",
      "epic 37\n",
      "equally 10\n",
      "era 7\n",
      "especially 39\n",
      "esque 6\n",
      "ethereal 7\n",
      "even 112\n",
      "evening 16\n",
      "event 19\n",
      "ever 126\n",
      "every 62\n",
      "everyone 34\n",
      "everything 22\n",
      "evil 13\n",
      "evocative 29\n",
      "examination 15\n",
      "examines 8\n",
      "example 15\n",
      "exceedingly 7\n",
      "excellent 89\n",
      "excels 7\n",
      "exceptional 22\n",
      "exceptionally 8\n",
      "excitement 17\n",
      "excites 7\n",
      "exciting 46\n",
      "executed 28\n",
      "execution 6\n",
      "exercise 13\n",
      "exhilarating 22\n",
      "existence 6\n",
      "exotic 10\n",
      "expect 6\n",
      "expectation 17\n",
      "experience 95\n",
      "expert 6\n",
      "expertly 8\n",
      "exploitative 6\n",
      "exploration 12\n",
      "explores 7\n",
      "express 9\n",
      "expressive 6\n",
      "exquisite 43\n",
      "exterior 8\n",
      "extraordinarily 16\n",
      "extraordinary 50\n",
      "extreme 6\n",
      "extremely 19\n",
      "eye 52\n",
      "eyed 8\n",
      "fabulous 10\n",
      "face 30\n",
      "fact 13\n",
      "fairly 16\n",
      "fairy 8\n",
      "faith 7\n",
      "faithful 8\n",
      "fall 11\n",
      "familiar 10\n",
      "family 105\n",
      "fan 46\n",
      "fanciful 8\n",
      "fantastic 32\n",
      "fantasy 29\n",
      "far 51\n",
      "fare 25\n",
      "fascinating 114\n",
      "fashion 11\n",
      "fashioned 24\n",
      "fast 21\n",
      "fat 11\n",
      "father 7\n",
      "favorite 10\n",
      "fear 14\n",
      "fearlessness 9\n",
      "feast 13\n",
      "feat 17\n",
      "feature 66\n",
      "featuring 6\n",
      "feel 95\n",
      "feeling 52\n",
      "felt 14\n",
      "female 28\n",
      "ferrara 6\n",
      "festival 22\n",
      "fi 15\n",
      "fiercely 7\n",
      "fight 14\n",
      "filled 13\n",
      "filling 6\n",
      "film 1104\n",
      "filmed 6\n",
      "filming 6\n",
      "filmmaker 54\n",
      "filmmaking 61\n",
      "finally 25\n",
      "find 50\n",
      "fine 69\n",
      "finely 19\n",
      "finest 17\n",
      "finish 6\n",
      "fire 14\n",
      "first 117\n",
      "fish 9\n",
      "fit 14\n",
      "flair 20\n",
      "flash 7\n",
      "flat 12\n",
      "flavor 8\n",
      "flawed 7\n",
      "flawless 10\n",
      "flick 41\n",
      "flow 6\n",
      "fluid 8\n",
      "folk 5\n",
      "follow 5\n",
      "fondness 5\n",
      "fontaine 6\n",
      "food 16\n",
      "footage 8\n",
      "force 20\n",
      "forceful 9\n",
      "forcefully 5\n",
      "foreign 12\n",
      "forget 22\n",
      "form 28\n",
      "format 14\n",
      "formed 8\n",
      "formula 15\n",
      "found 16\n",
      "four 16\n",
      "fourth 6\n",
      "frailty 8\n",
      "frame 15\n",
      "franchise 8\n",
      "frank 6\n",
      "free 11\n",
      "french 40\n",
      "frenetic 6\n",
      "frequent 8\n",
      "frequently 10\n",
      "fresh 45\n",
      "freshness 16\n",
      "friendship 27\n",
      "front 8\n",
      "frozen 5\n",
      "fulfill 6\n",
      "fulfilling 5\n",
      "full 76\n",
      "fully 12\n",
      "fun 168\n",
      "funnier 11\n",
      "funniest 40\n",
      "funny 347\n",
      "fused 7\n",
      "future 20\n",
      "fuzzy 7\n",
      "gag 5\n",
      "gai 6\n",
      "game 32\n",
      "gang 5\n",
      "gangster 12\n",
      "gasp 5\n",
      "gasping 10\n",
      "gay 11\n",
      "gem 18\n",
      "gender 7\n",
      "general 7\n",
      "generosity 5\n",
      "generous 10\n",
      "genius 7\n",
      "genre 60\n",
      "gentle 30\n",
      "gently 13\n",
      "genuine 49\n",
      "genuinely 33\n",
      "gere 6\n",
      "get 60\n",
      "getting 6\n",
      "ghetto 10\n",
      "giddy 6\n",
      "gift 9\n",
      "gifted 19\n",
      "girl 26\n",
      "give 83\n",
      "given 11\n",
      "giving 8\n",
      "glimpse 13\n",
      "glorious 9\n",
      "gloriously 8\n",
      "glued 5\n",
      "go 49\n",
      "goal 7\n",
      "god 9\n",
      "godard 5\n",
      "goer 9\n",
      "going 28\n",
      "goo 6\n",
      "good 339\n",
      "goodfellas 7\n",
      "gorgeous 90\n",
      "gorgeously 12\n",
      "gosling 7\n",
      "got 11\n",
      "grace 40\n",
      "graceful 6\n",
      "gracefully 9\n",
      "gradually 15\n",
      "grand 30\n",
      "grandeur 18\n",
      "grant 24\n",
      "graphic 13\n",
      "great 220\n",
      "greater 5\n",
      "greatest 36\n",
      "greatly 5\n",
      "greatness 8\n",
      "greek 7\n",
      "grew 5\n",
      "grief 11\n",
      "griffith 7\n",
      "grip 8\n",
      "gripping 26\n",
      "gritty 9\n",
      "group 6\n",
      "grow 5\n",
      "growing 6\n",
      "grown 20\n",
      "guaranteed 14\n",
      "guessing 12\n",
      "guilt 8\n",
      "guilty 12\n",
      "gut 7\n",
      "guy 7\n",
      "gyllenhaal 14\n",
      "hair 8\n",
      "half 23\n",
      "hand 8\n",
      "handed 6\n",
      "handled 7\n",
      "handsome 8\n",
      "happiness 9\n",
      "happy 21\n",
      "hard 57\n",
      "haunting 19\n",
      "head 24\n",
      "headed 7\n",
      "heady 7\n",
      "heard 9\n",
      "heart 161\n",
      "heartbreaking 14\n",
      "hearted 31\n",
      "heartfelt 35\n",
      "heartwarming 29\n",
      "heated 6\n",
      "heaven 11\n",
      "heightened 6\n",
      "held 7\n",
      "help 17\n",
      "hennings 8\n",
      "herald 7\n",
      "hero 15\n",
      "herzog 9\n",
      "high 58\n",
      "highest 10\n",
      "highly 46\n",
      "hilarious 81\n",
      "hilariously 11\n",
      "historical 19\n",
      "historically 9\n",
      "history 55\n",
      "hit 18\n",
      "hitting 8\n",
      "hoffman 13\n",
      "hold 28\n",
      "holiday 17\n",
      "holly 9\n",
      "hollywood 48\n",
      "holm 10\n",
      "holocaust 8\n",
      "holographic 6\n",
      "home 14\n",
      "honest 55\n",
      "honesty 12\n",
      "hong 6\n",
      "honored 6\n",
      "hoot 7\n",
      "hope 11\n",
      "hopeful 16\n",
      "hopefully 6\n",
      "hoping 6\n",
      "horrifying 14\n",
      "horror 30\n",
      "host 11\n",
      "hot 8\n",
      "hour 37\n",
      "house 7\n",
      "however 9\n",
      "huge 16\n",
      "hugely 30\n",
      "human 79\n",
      "humanity 17\n",
      "humor 130\n",
      "humorous 15\n",
      "humour 9\n",
      "hunter 9\n",
      "huppert 7\n",
      "hypnotic 7\n",
      "ian 6\n",
      "ice 8\n",
      "idea 42\n",
      "idealistic 6\n",
      "identity 10\n",
      "ii 6\n",
      "illuminating 7\n",
      "image 38\n",
      "imagery 13\n",
      "imagination 38\n",
      "imaginative 26\n",
      "imaginatively 8\n",
      "imagine 15\n",
      "imax 22\n",
      "immensely 20\n",
      "impact 13\n",
      "impeccable 14\n",
      "important 25\n",
      "impossible 22\n",
      "impressed 7\n",
      "impression 9\n",
      "impressive 47\n",
      "improbable 10\n",
      "improves 7\n",
      "improvised 6\n",
      "impulse 8\n",
      "including 11\n",
      "incredible 8\n",
      "incredibly 18\n",
      "indian 11\n",
      "indie 6\n",
      "inducing 6\n",
      "industry 6\n",
      "inescapably 7\n",
      "infectious 7\n",
      "informative 11\n",
      "ingenious 19\n",
      "inimitable 9\n",
      "injects 6\n",
      "innocence 8\n",
      "insane 7\n",
      "insanely 11\n",
      "inside 21\n",
      "insight 20\n",
      "insightful 21\n",
      "insightfully 7\n",
      "inspirational 13\n",
      "inspired 10\n",
      "inspiring 40\n",
      "instantly 9\n",
      "instead 6\n",
      "instinct 6\n",
      "integrity 6\n",
      "intelligence 64\n",
      "intelligent 76\n",
      "intelligently 12\n",
      "intended 11\n",
      "intense 37\n",
      "intensity 8\n",
      "interest 11\n",
      "interesting 57\n",
      "interestingly 8\n",
      "international 13\n",
      "interview 8\n",
      "intimacy 11\n",
      "intimate 30\n",
      "intoxicating 16\n",
      "intricate 12\n",
      "intrigue 12\n",
      "intriguing 44\n",
      "introduce 6\n",
      "introduction 6\n",
      "inuit 6\n",
      "invaluable 6\n",
      "inventive 66\n",
      "inventiveness 10\n",
      "inviting 7\n",
      "involved 9\n",
      "involving 11\n",
      "irresistible 19\n",
      "jackson 12\n",
      "jacques 6\n",
      "jake 7\n",
      "james 12\n",
      "japanese 11\n",
      "jaw 17\n",
      "jazzy 11\n",
      "jealousy 7\n",
      "jessica 9\n",
      "job 31\n",
      "john 17\n",
      "joke 14\n",
      "jolt 6\n",
      "jones 15\n",
      "journey 26\n",
      "joy 20\n",
      "joyous 29\n",
      "jumble 6\n",
      "jump 9\n",
      "junior 12\n",
      "justice 7\n",
      "kaufman 7\n",
      "keep 69\n",
      "key 15\n",
      "kid 70\n",
      "kidman 9\n",
      "killer 7\n",
      "kind 76\n",
      "kindness 6\n",
      "kinetic 7\n",
      "kinetically 7\n",
      "king 7\n",
      "kissing 9\n",
      "kline 8\n",
      "know 35\n",
      "knowledge 9\n",
      "known 14\n",
      "kong 6\n",
      "korean 11\n",
      "krawczyk 6\n",
      "kwan 12\n",
      "lacking 6\n",
      "landscape 21\n",
      "lane 15\n",
      "large 20\n",
      "larger 6\n",
      "last 48\n",
      "laugh 99\n",
      "laughed 8\n",
      "laughing 7\n",
      "laughter 20\n",
      "lavish 6\n",
      "layer 8\n",
      "layered 11\n",
      "le 15\n",
      "lead 31\n",
      "leaf 12\n",
      "leap 6\n",
      "least 30\n",
      "leave 26\n",
      "leaving 13\n",
      "lee 9\n",
      "left 8\n",
      "leguizamo 10\n",
      "lends 11\n",
      "lesson 8\n",
      "let 7\n",
      "level 29\n",
      "librarian 6\n",
      "lie 7\n",
      "life 183\n",
      "lift 18\n",
      "light 24\n",
      "like 151\n",
      "likeable 12\n",
      "likely 11\n",
      "line 18\n",
      "list 10\n",
      "literary 9\n",
      "literate 6\n",
      "little 85\n",
      "live 14\n",
      "lived 14\n",
      "lively 10\n",
      "living 9\n",
      "locale 17\n",
      "location 13\n",
      "lofty 11\n",
      "long 40\n",
      "look 100\n",
      "looking 27\n",
      "loss 7\n",
      "lost 9\n",
      "lot 56\n",
      "loud 9\n",
      "love 202\n",
      "loved 17\n",
      "lovely 25\n",
      "lover 12\n",
      "low 18\n",
      "lrb 137\n",
      "luminous 6\n",
      "lunacy 6\n",
      "lush 12\n",
      "lustrous 6\n",
      "lyrical 12\n",
      "machine 6\n",
      "made 141\n",
      "madness 6\n",
      "magic 22\n",
      "magical 19\n",
      "magnetic 6\n",
      "magnificent 28\n",
      "mainstream 10\n",
      "maintain 7\n",
      "majestic 13\n",
      "major 7\n",
      "make 249\n",
      "maker 7\n",
      "making 50\n",
      "mamet 8\n",
      "man 59\n",
      "manage 7\n",
      "manages 45\n",
      "manhood 10\n",
      "manner 19\n",
      "many 97\n",
      "mark 20\n",
      "martha 7\n",
      "martial 7\n",
      "martin 6\n",
      "marvelous 24\n",
      "marvelously 11\n",
      "massive 7\n",
      "master 32\n",
      "masterful 40\n",
      "masterfully 20\n",
      "masterpiece 63\n",
      "material 20\n",
      "matter 12\n",
      "mature 8\n",
      "maturity 7\n",
      "may 77\n",
      "mean 6\n",
      "meaning 10\n",
      "meaningful 7\n",
      "measure 9\n",
      "measured 5\n",
      "meditation 11\n",
      "meditative 5\n",
      "medium 12\n",
      "mel 5\n",
      "melancholy 5\n",
      "melodrama 21\n",
      "member 9\n",
      "memorable 46\n",
      "memory 22\n",
      "men 21\n",
      "mention 14\n",
      "mercy 10\n",
      "mesmerize 5\n",
      "mesmerizing 42\n",
      "message 29\n",
      "metaphor 7\n",
      "metropolis 7\n",
      "michael 15\n",
      "michel 6\n",
      "middle 19\n",
      "might 36\n",
      "mike 8\n",
      "mind 40\n",
      "minimalist 6\n",
      "minor 11\n",
      "minority 7\n",
      "minute 33\n",
      "mira 7\n",
      "miracle 6\n",
      "miraculous 7\n",
      "mix 14\n",
      "mixed 8\n",
      "mixture 10\n",
      "miyazaki 7\n",
      "modern 33\n",
      "modest 8\n",
      "moment 59\n",
      "monster 14\n",
      "month 6\n",
      "monty 8\n",
      "monumental 8\n",
      "mood 22\n",
      "moore 15\n",
      "moral 15\n",
      "mostly 17\n",
      "mother 19\n",
      "motion 34\n",
      "motivated 7\n",
      "mouse 6\n",
      "mouth 9\n",
      "move 14\n",
      "movie 757\n",
      "moviegoing 5\n",
      "moviemaking 14\n",
      "moving 131\n",
      "mr 27\n",
      "much 126\n",
      "multi 5\n",
      "multilayered 5\n",
      "multiple 5\n",
      "murder 8\n",
      "murphy 8\n",
      "music 81\n",
      "musical 11\n",
      "musician 8\n",
      "must 30\n",
      "mysterious 10\n",
      "mystery 23\n",
      "mystical 5\n",
      "mythic 6\n",
      "nair 9\n",
      "name 7\n",
      "namesake 6\n",
      "narrative 33\n",
      "narrator 5\n",
      "national 8\n",
      "natural 23\n",
      "naturalistic 5\n",
      "naturally 10\n",
      "nature 19\n",
      "natured 8\n",
      "near 11\n",
      "nearly 22\n",
      "need 24\n",
      "nerve 10\n",
      "never 98\n",
      "new 115\n",
      "newcomer 20\n",
      "news 8\n",
      "next 18\n",
      "nice 18\n",
      "nicely 28\n",
      "niche 8\n",
      "nicholson 10\n",
      "night 6\n",
      "niro 6\n",
      "noble 5\n",
      "noir 6\n",
      "nominated 10\n",
      "nomination 19\n",
      "non 16\n",
      "nonchalant 5\n",
      "nonstop 7\n",
      "nostalgic 6\n",
      "notch 25\n",
      "note 8\n",
      "noteworthy 7\n",
      "nothing 32\n",
      "nourishing 5\n",
      "novel 12\n",
      "noyce 6\n",
      "nuance 7\n",
      "number 14\n",
      "observation 16\n",
      "obvious 5\n",
      "occasionally 18\n",
      "odd 13\n",
      "oddly 8\n",
      "odds 10\n",
      "ode 9\n",
      "offbeat 22\n",
      "offer 43\n",
      "offering 8\n",
      "often 93\n",
      "old 57\n",
      "one 517\n",
      "open 13\n",
      "openness 8\n",
      "opera 18\n",
      "opportunity 5\n",
      "order 11\n",
      "ordinary 7\n",
      "oriented 12\n",
      "original 65\n",
      "originality 34\n",
      "orlando 6\n",
      "oscar 62\n",
      "others 9\n",
      "otherwise 12\n",
      "outer 8\n",
      "outrageous 13\n",
      "outstanding 37\n",
      "overall 17\n",
      "overdoing 5\n",
      "pace 25\n",
      "paced 16\n",
      "pacino 9\n",
      "pack 7\n",
      "package 12\n",
      "packed 24\n",
      "page 9\n",
      "painful 9\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "paint 6\n",
      "painting 12\n",
      "pair 18\n",
      "palette 13\n",
      "panache 6\n",
      "panoramic 6\n",
      "pantheon 5\n",
      "parable 7\n",
      "paradiso 9\n",
      "paranoia 6\n",
      "parent 24\n",
      "part 55\n",
      "particular 5\n",
      "particularly 10\n",
      "party 7\n",
      "passion 44\n",
      "passionate 10\n",
      "past 12\n",
      "pat 6\n",
      "path 5\n",
      "pathology 9\n",
      "pathos 7\n",
      "paul 5\n",
      "paxton 10\n",
      "people 55\n",
      "perceptive 5\n",
      "perfect 110\n",
      "perfected 6\n",
      "perfection 13\n",
      "perfectly 42\n",
      "performance 472\n",
      "performed 6\n",
      "performer 11\n",
      "perhaps 5\n",
      "peril 7\n",
      "period 18\n",
      "person 7\n",
      "persona 8\n",
      "personal 13\n",
      "perspective 9\n",
      "pertinent 7\n",
      "peter 8\n",
      "phenomenal 20\n",
      "philosophical 5\n",
      "photographed 12\n",
      "photography 11\n",
      "physical 8\n",
      "pianist 8\n",
      "pic 12\n",
      "piccoli 5\n",
      "picture 99\n",
      "piece 100\n",
      "pitch 15\n",
      "place 41\n",
      "planet 7\n",
      "play 22\n",
      "played 9\n",
      "playful 18\n",
      "playing 7\n",
      "pleasant 10\n",
      "please 7\n",
      "pleaser 14\n",
      "pleasing 13\n",
      "pleasurable 18\n",
      "pleasure 42\n",
      "plenty 16\n",
      "plot 21\n",
      "plummer 7\n",
      "poem 16\n",
      "poetic 8\n",
      "poetry 15\n",
      "poignancy 11\n",
      "poignant 19\n",
      "point 17\n",
      "polanski 11\n",
      "polished 21\n",
      "political 20\n",
      "politically 11\n",
      "pompeo 9\n",
      "pop 11\n",
      "popcorn 12\n",
      "popular 7\n",
      "portrait 49\n",
      "portrayal 19\n",
      "positive 11\n",
      "possible 25\n",
      "possibly 20\n",
      "potent 15\n",
      "potential 9\n",
      "pounding 8\n",
      "power 63\n",
      "powerful 105\n",
      "powerfully 6\n",
      "predecessor 13\n",
      "predict 7\n",
      "premise 17\n",
      "presence 7\n",
      "present 12\n",
      "pretty 40\n",
      "prevention 7\n",
      "previous 8\n",
      "price 9\n",
      "primarily 6\n",
      "principal 6\n",
      "private 9\n",
      "privileged 6\n",
      "probably 25\n",
      "process 10\n",
      "produced 10\n",
      "production 22\n",
      "prof 30\n",
      "professional 14\n",
      "profile 9\n",
      "profound 14\n",
      "project 15\n",
      "promise 14\n",
      "protagonist 14\n",
      "proud 15\n",
      "provide 7\n",
      "provides 15\n",
      "providing 6\n",
      "provocative 26\n",
      "provoking 26\n",
      "pseudo 6\n",
      "psychological 35\n",
      "psychologically 6\n",
      "psychology 6\n",
      "pull 20\n",
      "punch 6\n",
      "punny 6\n",
      "pure 24\n",
      "purely 16\n",
      "purpose 6\n",
      "put 35\n",
      "quality 16\n",
      "question 21\n",
      "quiet 20\n",
      "quirky 41\n",
      "quite 54\n",
      "quivering 7\n",
      "race 14\n",
      "rachel 8\n",
      "radiant 6\n",
      "raise 10\n",
      "ralph 8\n",
      "ramsay 10\n",
      "rank 25\n",
      "rapid 6\n",
      "rapt 6\n",
      "rare 61\n",
      "rarely 10\n",
      "rate 36\n",
      "rather 44\n",
      "rattling 6\n",
      "ravishing 12\n",
      "raw 8\n",
      "reacting 10\n",
      "read 17\n",
      "reaffirming 7\n",
      "real 76\n",
      "realism 7\n",
      "realistic 15\n",
      "reality 11\n",
      "realized 19\n",
      "really 92\n",
      "reason 12\n",
      "receives 9\n",
      "recent 51\n",
      "recipe 9\n",
      "recommend 18\n",
      "recommendation 7\n",
      "recovery 6\n",
      "red 9\n",
      "rediscover 7\n",
      "reef 9\n",
      "refreshes 9\n",
      "refreshing 24\n",
      "refreshingly 20\n",
      "regret 6\n",
      "relationship 30\n",
      "relaxed 7\n",
      "release 19\n",
      "released 6\n",
      "remains 24\n",
      "remake 8\n",
      "remarkable 78\n",
      "remarkably 29\n",
      "remember 10\n",
      "reminded 6\n",
      "reminder 8\n",
      "reminds 6\n",
      "rendered 11\n",
      "renowned 8\n",
      "report 8\n",
      "represents 7\n",
      "resist 7\n",
      "resonant 14\n",
      "resourceful 8\n",
      "respect 17\n",
      "rest 17\n",
      "result 15\n",
      "return 16\n",
      "revealing 15\n",
      "reveals 22\n",
      "revelation 8\n",
      "revelatory 10\n",
      "revenge 6\n",
      "revisiting 6\n",
      "reward 8\n",
      "rewarded 7\n",
      "rewarding 15\n",
      "rhythm 15\n",
      "rich 58\n",
      "richer 12\n",
      "richly 26\n",
      "ride 37\n",
      "right 57\n",
      "ring 8\n",
      "ripe 6\n",
      "rise 7\n",
      "rising 7\n",
      "rival 6\n",
      "riveting 42\n",
      "road 11\n",
      "robert 10\n",
      "rock 25\n",
      "role 39\n",
      "roll 17\n",
      "rollicking 7\n",
      "romance 39\n",
      "romantic 84\n",
      "root 5\n",
      "rousing 5\n",
      "rrb 146\n",
      "run 13\n",
      "rural 5\n",
      "russian 6\n",
      "ryan 10\n",
      "sad 15\n",
      "sadness 7\n",
      "said 6\n",
      "san 5\n",
      "sandler 9\n",
      "sane 12\n",
      "sarandon 11\n",
      "satire 8\n",
      "satisfied 6\n",
      "satisfies 9\n",
      "satisfying 79\n",
      "savory 10\n",
      "savvy 10\n",
      "saw 9\n",
      "say 21\n",
      "scale 10\n",
      "scary 7\n",
      "scenario 6\n",
      "scene 60\n",
      "scenery 5\n",
      "school 5\n",
      "sci 15\n",
      "scientist 10\n",
      "scooping 8\n",
      "scope 9\n",
      "score 31\n",
      "scorsese 7\n",
      "scratch 11\n",
      "screen 94\n",
      "screenplay 41\n",
      "screenwriter 6\n",
      "screenwriting 5\n",
      "script 40\n",
      "sea 8\n",
      "searing 6\n",
      "season 18\n",
      "seat 45\n",
      "second 11\n",
      "secret 11\n",
      "see 128\n",
      "seeing 27\n",
      "seek 7\n",
      "seeking 5\n",
      "seem 22\n",
      "seems 17\n",
      "seen 74\n",
      "self 45\n",
      "sensational 9\n",
      "sense 78\n",
      "sensitive 26\n",
      "sensual 22\n",
      "sentimental 16\n",
      "sequel 29\n",
      "sequence 45\n",
      "series 9\n",
      "serious 16\n",
      "served 8\n",
      "set 30\n",
      "setting 8\n",
      "seven 6\n",
      "several 6\n",
      "sex 22\n",
      "sexual 11\n",
      "sexy 27\n",
      "shape 12\n",
      "shaped 11\n",
      "share 12\n",
      "sharp 18\n",
      "sheer 15\n",
      "sheerly 6\n",
      "shifting 9\n",
      "shimmering 9\n",
      "shine 13\n",
      "shining 9\n",
      "shiver 6\n",
      "shock 8\n",
      "shocking 6\n",
      "short 25\n",
      "shot 51\n",
      "show 47\n",
      "showtime 6\n",
      "shyamalan 6\n",
      "sick 11\n",
      "side 26\n",
      "sight 12\n",
      "sign 9\n",
      "significance 7\n",
      "significant 9\n",
      "significantly 6\n",
      "silly 17\n",
      "silver 6\n",
      "simple 35\n",
      "simply 35\n",
      "simultaneously 17\n",
      "since 50\n",
      "sincere 17\n",
      "sincerity 6\n",
      "singer 7\n",
      "singing 18\n",
      "single 7\n",
      "situation 14\n",
      "skilfully 6\n",
      "skill 32\n",
      "skillful 9\n",
      "skillfully 10\n",
      "skin 10\n",
      "sleepless 6\n",
      "slice 11\n",
      "slightly 6\n",
      "slowly 6\n",
      "small 29\n",
      "smart 106\n",
      "smarter 6\n",
      "smartly 19\n",
      "smile 22\n",
      "sneak 9\n",
      "soap 6\n",
      "soar 18\n",
      "sobering 9\n",
      "social 14\n",
      "soderbergh 9\n",
      "solid 106\n",
      "solidly 15\n",
      "somber 6\n",
      "somehow 8\n",
      "someone 6\n",
      "something 90\n",
      "sometimes 39\n",
      "somewhat 8\n",
      "son 13\n",
      "song 10\n",
      "soon 7\n",
      "sophisticated 20\n",
      "sort 15\n",
      "soul 27\n",
      "sound 16\n",
      "soundtrack 12\n",
      "space 19\n",
      "spader 7\n",
      "spare 13\n",
      "sparkling 16\n",
      "special 44\n",
      "spectacle 9\n",
      "spectacular 19\n",
      "spectacularly 12\n",
      "spell 10\n",
      "spellbinding 9\n",
      "spider 15\n",
      "spielberg 26\n",
      "spike 9\n",
      "spin 12\n",
      "spirit 64\n",
      "spirited 15\n",
      "spiritual 16\n",
      "spirituality 6\n",
      "splendid 36\n",
      "splendor 9\n",
      "spooky 8\n",
      "sport 11\n",
      "sportsman 8\n",
      "spy 29\n",
      "stable 6\n",
      "stage 9\n",
      "staggeringly 6\n",
      "stand 27\n",
      "standard 9\n",
      "standout 8\n",
      "stanley 10\n",
      "star 63\n",
      "starring 6\n",
      "start 10\n",
      "started 10\n",
      "startling 10\n",
      "station 8\n",
      "stay 13\n",
      "steal 12\n",
      "stein 6\n",
      "stellar 7\n",
      "step 8\n",
      "stepped 6\n",
      "steven 12\n",
      "still 84\n",
      "stirring 19\n",
      "stop 11\n",
      "story 280\n",
      "storyline 17\n",
      "storytelling 36\n",
      "straightforward 7\n",
      "strange 13\n",
      "strangely 6\n",
      "strength 9\n",
      "stretch 9\n",
      "strike 11\n",
      "striking 21\n",
      "string 12\n",
      "strong 52\n",
      "strongest 10\n",
      "structured 6\n",
      "struggle 14\n",
      "struggled 10\n",
      "studio 8\n",
      "study 37\n",
      "stuff 29\n",
      "stuffiest 8\n",
      "stunning 54\n",
      "style 50\n",
      "stylish 40\n",
      "stylist 10\n",
      "subculture 8\n",
      "subject 44\n",
      "sublime 9\n",
      "sublimely 12\n",
      "substantial 9\n",
      "subtitle 7\n",
      "subtle 32\n",
      "subtlest 9\n",
      "subversive 6\n",
      "succeeded 6\n",
      "succeeds 12\n",
      "success 24\n",
      "successful 37\n",
      "suck 7\n",
      "suffering 6\n",
      "sugar 6\n",
      "summer 56\n",
      "sumptuous 6\n",
      "sun 8\n",
      "sundance 6\n",
      "superb 38\n",
      "superbly 13\n",
      "superior 24\n",
      "supply 6\n",
      "supporting 7\n",
      "supremely 7\n",
      "sure 16\n",
      "surefire 7\n",
      "surprise 25\n",
      "surprising 43\n",
      "surprisingly 61\n",
      "survivable 6\n",
      "susan 9\n",
      "suspense 30\n",
      "suspenseful 22\n",
      "sustained 10\n",
      "sway 6\n",
      "sweeping 10\n",
      "sweet 109\n",
      "sweetly 16\n",
      "sweetness 11\n",
      "swept 8\n",
      "swooping 15\n",
      "sympathetic 13\n",
      "sympathy 6\n",
      "tackle 6\n",
      "take 81\n",
      "taken 7\n",
      "taking 12\n",
      "tale 89\n",
      "talent 53\n",
      "talented 36\n",
      "talking 7\n",
      "tasteful 7\n",
      "teacher 8\n",
      "team 9\n",
      "tear 28\n",
      "technical 25\n",
      "technically 10\n",
      "technology 17\n",
      "teeming 13\n",
      "teenager 6\n",
      "tell 28\n",
      "telling 10\n",
      "tender 9\n",
      "tense 10\n",
      "tension 13\n",
      "term 12\n",
      "terrific 94\n",
      "terrifying 6\n",
      "territory 8\n",
      "testud 7\n",
      "th 9\n",
      "thanks 30\n",
      "theater 28\n",
      "theatre 8\n",
      "theme 31\n",
      "therapy 6\n",
      "thing 53\n",
      "think 6\n",
      "thinking 17\n",
      "thoroughly 33\n",
      "though 24\n",
      "thought 52\n",
      "thoughtful 50\n",
      "thoughtfully 8\n",
      "three 35\n",
      "thrill 26\n",
      "thriller 108\n",
      "throat 9\n",
      "throughout 23\n",
      "thumb 9\n",
      "tickle 7\n",
      "time 182\n",
      "timeless 13\n",
      "timing 11\n",
      "tiny 10\n",
      "title 14\n",
      "today 21\n",
      "together 39\n",
      "told 23\n",
      "tone 32\n",
      "tool 6\n",
      "top 54\n",
      "totally 6\n",
      "touch 34\n",
      "touched 8\n",
      "touching 86\n",
      "tough 7\n",
      "tour 12\n",
      "town 6\n",
      "trademark 6\n",
      "tradition 7\n",
      "traditional 8\n",
      "tragedy 11\n",
      "tragic 7\n",
      "transcend 6\n",
      "transcendent 5\n",
      "transcends 13\n",
      "transform 8\n",
      "transformed 6\n",
      "transforms 8\n",
      "treasure 17\n",
      "treat 34\n",
      "treatment 8\n",
      "tree 6\n",
      "trek 5\n",
      "tremendous 16\n",
      "tremendously 5\n",
      "tribute 18\n",
      "trick 5\n",
      "trip 11\n",
      "triumph 29\n",
      "triumphant 12\n",
      "true 43\n",
      "truly 50\n",
      "truth 9\n",
      "try 7\n",
      "trying 7\n",
      "tsai 10\n",
      "tug 6\n",
      "turn 56\n",
      "turned 6\n",
      "turning 10\n",
      "twice 16\n",
      "twist 15\n",
      "twisted 28\n",
      "twisty 6\n",
      "two 69\n",
      "ultimate 24\n",
      "ultimately 47\n",
      "unabashedly 13\n",
      "uncle 8\n",
      "uncommonly 11\n",
      "unconventional 8\n",
      "undeniable 13\n",
      "undeniably 20\n",
      "understand 9\n",
      "understands 8\n",
      "understated 8\n",
      "unexpected 19\n",
      "unexpectedly 10\n",
      "unfolding 6\n",
      "unfolds 15\n",
      "unforced 8\n",
      "unforgettable 7\n",
      "uniformly 6\n",
      "unique 26\n",
      "universal 28\n",
      "unlike 9\n",
      "unlikely 7\n",
      "unpredictable 33\n",
      "unsentimental 7\n",
      "unsettling 22\n",
      "unusual 7\n",
      "updating 6\n",
      "uplifting 17\n",
      "upon 7\n",
      "upper 6\n",
      "ups 8\n",
      "urban 11\n",
      "urgently 10\n",
      "us 24\n",
      "use 28\n",
      "usual 15\n",
      "usually 8\n",
      "utterly 24\n",
      "value 17\n",
      "van 6\n",
      "variation 6\n",
      "variety 6\n",
      "vastly 6\n",
      "vehicle 6\n",
      "vein 6\n",
      "venice 6\n",
      "veracity 7\n",
      "verbal 6\n",
      "version 17\n",
      "verve 12\n",
      "veteran 7\n",
      "vibrant 30\n",
      "video 12\n",
      "view 16\n",
      "viewer 26\n",
      "viewing 19\n",
      "villain 20\n",
      "vintage 8\n",
      "visceral 16\n",
      "vision 16\n",
      "visual 76\n",
      "visually 61\n",
      "visuals 25\n",
      "vital 15\n",
      "vivid 37\n",
      "vividly 16\n",
      "voice 6\n",
      "vulgar 10\n",
      "walk 8\n",
      "wall 15\n",
      "wallop 8\n",
      "want 28\n",
      "wanting 8\n",
      "war 35\n",
      "warm 47\n",
      "warmth 44\n",
      "washington 27\n",
      "watch 67\n",
      "watchable 20\n",
      "watching 37\n",
      "water 18\n",
      "wattage 8\n",
      "wave 15\n",
      "way 123\n",
      "web 7\n",
      "wedding 15\n",
      "week 6\n",
      "welcome 21\n",
      "well 317\n",
      "whale 9\n",
      "whether 7\n",
      "whimsical 15\n",
      "whimsy 7\n",
      "white 16\n",
      "whole 35\n",
      "wholesome 5\n",
      "wholly 5\n",
      "whose 32\n",
      "wicked 6\n",
      "wickedly 27\n",
      "wide 15\n",
      "wild 17\n",
      "wildly 13\n",
      "william 6\n",
      "williams 9\n",
      "willing 10\n",
      "win 25\n",
      "wind 10\n",
      "winner 25\n",
      "winning 41\n",
      "wisdom 9\n",
      "wise 16\n",
      "wish 6\n",
      "wit 44\n",
      "within 5\n",
      "without 77\n",
      "witty 37\n",
      "wo 19\n",
      "woman 57\n",
      "wonder 27\n",
      "wonderful 82\n",
      "wonderfully 34\n",
      "wonderous 9\n",
      "wondrous 14\n",
      "wondrously 14\n",
      "woody 10\n",
      "word 21\n",
      "work 274\n",
      "working 28\n",
      "world 108\n",
      "worry 10\n",
      "worth 76\n",
      "worthwhile 28\n",
      "worthy 49\n",
      "would 30\n",
      "woven 8\n",
      "wrapped 6\n",
      "wrenching 12\n",
      "writer 31\n",
      "writing 28\n",
      "written 62\n",
      "wrought 7\n",
      "wry 12\n",
      "ya 8\n",
      "yarn 17\n",
      "year 275\n",
      "yes 5\n",
      "yesterday 8\n",
      "yet 48\n",
      "yiddish 6\n",
      "york 15\n",
      "young 55\n",
      "zing 11\n",
      "zinger 5\n",
      "zippy 10\n"
     ]
    }
   ],
   "source": [
    "dist = np.sum(pos_words, axis=0)\n",
    "for tag, count in zip(pos, dist):\n",
    "    print (tag,count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "postive_new= pd.DataFrame(dist)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "postive_new.columns=['word_count']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "postive_new['word'] = pd.Series(pos, index=postive_new.index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "postive_new1=postive_new[['word','word_count']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "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>word</th>\n",
       "      <th>word_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ability</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>able</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>absolute</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>absolutely</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>absorbing</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         word  word_count\n",
       "0     ability          20\n",
       "1        able          32\n",
       "2    absolute           9\n",
       "3  absolutely          24\n",
       "4   absorbing          18"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "postive_new1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "top_30_words=postive_new1.sort_values(['word_count'],ascending=[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "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>word</th>\n",
       "      <th>word_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>652</th>\n",
       "      <td>film</td>\n",
       "      <td>1104</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1149</th>\n",
       "      <td>movie</td>\n",
       "      <td>757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1223</th>\n",
       "      <td>one</td>\n",
       "      <td>517</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1280</th>\n",
       "      <td>performance</td>\n",
       "      <td>472</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>161</th>\n",
       "      <td>best</td>\n",
       "      <td>394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>715</th>\n",
       "      <td>funny</td>\n",
       "      <td>347</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>760</th>\n",
       "      <td>good</td>\n",
       "      <td>339</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1933</th>\n",
       "      <td>well</td>\n",
       "      <td>317</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1672</th>\n",
       "      <td>story</td>\n",
       "      <td>280</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1990</th>\n",
       "      <td>year</td>\n",
       "      <td>275</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1972</th>\n",
       "      <td>work</td>\n",
       "      <td>274</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>315</th>\n",
       "      <td>comedy</td>\n",
       "      <td>271</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1067</th>\n",
       "      <td>make</td>\n",
       "      <td>249</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>774</th>\n",
       "      <td>great</td>\n",
       "      <td>220</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>262</th>\n",
       "      <td>character</td>\n",
       "      <td>215</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1045</th>\n",
       "      <td>love</td>\n",
       "      <td>202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1020</th>\n",
       "      <td>life</td>\n",
       "      <td>183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1789</th>\n",
       "      <td>time</td>\n",
       "      <td>182</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>552</th>\n",
       "      <td>entertaining</td>\n",
       "      <td>168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>712</th>\n",
       "      <td>fun</td>\n",
       "      <td>168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>811</th>\n",
       "      <td>heart</td>\n",
       "      <td>161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1023</th>\n",
       "      <td>like</td>\n",
       "      <td>151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>476</th>\n",
       "      <td>drama</td>\n",
       "      <td>147</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1483</th>\n",
       "      <td>rrb</td>\n",
       "      <td>146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>244</th>\n",
       "      <td>cast</td>\n",
       "      <td>146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>458</th>\n",
       "      <td>director</td>\n",
       "      <td>143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1057</th>\n",
       "      <td>made</td>\n",
       "      <td>141</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1050</th>\n",
       "      <td>lrb</td>\n",
       "      <td>137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1152</th>\n",
       "      <td>moving</td>\n",
       "      <td>131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>actor</td>\n",
       "      <td>130</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              word  word_count\n",
       "652           film        1104\n",
       "1149         movie         757\n",
       "1223           one         517\n",
       "1280   performance         472\n",
       "161           best         394\n",
       "715          funny         347\n",
       "760           good         339\n",
       "1933          well         317\n",
       "1672         story         280\n",
       "1990          year         275\n",
       "1972          work         274\n",
       "315         comedy         271\n",
       "1067          make         249\n",
       "774          great         220\n",
       "262      character         215\n",
       "1045          love         202\n",
       "1020          life         183\n",
       "1789          time         182\n",
       "552   entertaining         168\n",
       "712            fun         168\n",
       "811          heart         161\n",
       "1023          like         151\n",
       "476          drama         147\n",
       "1483           rrb         146\n",
       "244           cast         146\n",
       "458       director         143\n",
       "1057          made         141\n",
       "1050           lrb         137\n",
       "1152        moving         131\n",
       "21           actor         130"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "top_30_words.head(30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "cv = CountVectorizer(max_features = 1500)\n",
    "x__train = cv.fit_transform(corpus).toarray()\n",
    "x__test= cv.fit_transform(corpus1).toarray()\n",
    "y = train_data.iloc[:, 2].values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Time to predict"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## using multinomial Naive Bayes due it is go good for discrete features , in our case we have 5 features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/song-yeongsug/.local/lib/python3.6/site-packages/sklearn/cross_validation.py:41: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.\n",
      "  \"This module will be removed in 0.20.\", DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "from sklearn.cross_validation import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(x__train, y, test_size = 0.40, random_state = 0)\n",
    "\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "classifier = MultinomialNB()\n",
    "classifier.fit(X_train, y_train)\n",
    "\n",
    "y_pred = classifier.predict(X_test)\n",
    "y_real_pred = classifier.predict(x__test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "mse = ((y_pred - y_test) ** 2).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6473792131231577"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#mean squre error\n",
    "mse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "rmse = sqrt(mse)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8045987901576522"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "코드 참고 : https://github.com/aptlo10/-Sentiment-Analysis-on-Movie-Reviews/blob/master/Sentiment_Analysison_MovieReviews_15-2-2018.ipynb"
   ]
  }
 ],
 "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}