twitter_sentiment_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/kaggle/input/test.csv\n",
      "/kaggle/input/train.csv\n"
     ]
    }
   ],
   "source": [
    "import numpy as np \n",
    "import pandas as pd \n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import seaborn as sns\n",
    "df1 = pd.read_csv('train.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0",
    "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a"
   },
   "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>id</th>\n",
       "      <th>label</th>\n",
       "      <th>tweet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>@user when a father is dysfunctional and is s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>@user @user thanks for #lyft credit i can't us...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>bihday your majesty</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>#model   i love u take with u all the time in ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>factsguide: society now    #motivation</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id  label                                              tweet\n",
       "0   1      0   @user when a father is dysfunctional and is s...\n",
       "1   2      0  @user @user thanks for #lyft credit i can't us...\n",
       "2   3      0                                bihday your majesty\n",
       "3   4      0  #model   i love u take with u all the time in ...\n",
       "4   5      0             factsguide: society now    #motivation"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id       0\n",
       "label    0\n",
       "tweet    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "29720 Non-Hatred Tweets\n",
      "2242 Hatred Tweets\n"
     ]
    }
   ],
   "source": [
    "print(len(df1[df1.label == 0]), 'Non-Hatred Tweets')\n",
    "print(len(df1[df1.label == 1]), 'Hatred Tweets') "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Preprocessing**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import nltk\n",
    "from sklearn import re \n",
    "from nltk.corpus import stopwords\n",
    "from nltk.stem import WordNetLemmatizer\n",
    "lemma = WordNetLemmatizer()\n",
    "from wordcloud import WordCloud, STOPWORDS\n",
    "from nltk import FreqDist \n",
    "\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import confusion_matrix, classification_report, f1_score "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Preprocessing the tweet column**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def normalizer(tweet):\n",
    "    tweets = \" \".join(filter(lambda x: x[0]!= '@' , tweet.split()))\n",
    "    tweets = re.sub('[^a-zA-Z]', ' ', tweets)\n",
    "    tweets = tweets.lower()\n",
    "    tweets = tweets.split()\n",
    "    tweets = [word for word in tweets if not word in set(stopwords.words('english'))]\n",
    "    tweets = [lemma.lemmatize(word) for word in tweets]\n",
    "    tweets = \" \".join(tweets)\n",
    "    return tweets\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1['normalized_text'] = df1.tweet.apply(normalizer)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Extracting words with hashtag for further analysis**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def extract_hashtag(tweet):\n",
    "    tweets = \" \".join(filter(lambda x: x[0]== '#', tweet.split()))\n",
    "    tweets = re.sub('[^a-zA-Z]',' ',  tweets)\n",
    "    tweets = tweets.lower()\n",
    "    tweets = [lemma.lemmatize(word) for word in tweets]\n",
    "    tweets = \"\".join(tweets)\n",
    "    return tweets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1['hashtag'] = df1.tweet.apply(extract_hashtag)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>id</th>\n",
       "      <th>label</th>\n",
       "      <th>tweet</th>\n",
       "      <th>normalized_text</th>\n",
       "      <th>hashtag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>@user when a father is dysfunctional and is s...</td>\n",
       "      <td>father dysfunctional selfish drag kid dysfunct...</td>\n",
       "      <td>run</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>@user @user thanks for #lyft credit i can't us...</td>\n",
       "      <td>thanks lyft credit use cause offer wheelchair ...</td>\n",
       "      <td>lyft  disapointed  getthanked</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>bihday your majesty</td>\n",
       "      <td>bihday majesty</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>#model   i love u take with u all the time in ...</td>\n",
       "      <td>model love u take u time ur</td>\n",
       "      <td>model</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>factsguide: society now    #motivation</td>\n",
       "      <td>factsguide society motivation</td>\n",
       "      <td>motivation</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   id  label                                              tweet  \\\n",
       "0   1      0   @user when a father is dysfunctional and is s...   \n",
       "1   2      0  @user @user thanks for #lyft credit i can't us...   \n",
       "2   3      0                                bihday your majesty   \n",
       "3   4      0  #model   i love u take with u all the time in ...   \n",
       "4   5      0             factsguide: society now    #motivation   \n",
       "\n",
       "                                     normalized_text  \\\n",
       "0  father dysfunctional selfish drag kid dysfunct...   \n",
       "1  thanks lyft credit use cause offer wheelchair ...   \n",
       "2                                     bihday majesty   \n",
       "3                        model love u take u time ur   \n",
       "4                      factsguide society motivation   \n",
       "\n",
       "                          hashtag  \n",
       "0                             run  \n",
       "1   lyft  disapointed  getthanked  \n",
       "2                                  \n",
       "3                           model  \n",
       "4                      motivation  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**To Create Cloud of words for all words and hatred words**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# all tweets \n",
    "all_words = \" \".join(df1.normalized_text)\n",
    "#print(all_all_words)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Hatred tweets\n",
    "hatred_words = \" \".join(df1[df1['label']==1].normalized_text)\n",
    "#print(hatred_words)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**All tweets cloudword**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "wordcloud = WordCloud(height=2000, width=2000, stopwords=STOPWORDS, background_color='white')\n",
    "wordcloud = wordcloud.generate(all_words)\n",
    "plt.imshow(wordcloud)\n",
    "plt.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**hatred tweets cloudword**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "wordcloud = WordCloud(height=2000, width=2000, stopwords=STOPWORDS, background_color='white')\n",
    "wordcloud = wordcloud.generate(hatred_words)\n",
    "plt.imshow(wordcloud)\n",
    "plt.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Analysing Hashtag words**\n",
    "\n",
    "**plotting the most common hashtag used in tweets**\n",
    "**all hashtag \n",
    "hatred hashtag**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('love', 1591),\n",
       " ('positive', 879),\n",
       " ('smile', 577),\n",
       " ('healthy', 576),\n",
       " ('thankful', 471),\n",
       " ('fun', 463),\n",
       " ('life', 438),\n",
       " ('summer', 392),\n",
       " ('model', 366),\n",
       " ('cute', 364)]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "freq_all_hashtag = FreqDist(list((\" \".join(df1.hashtag)).split())).most_common(10)\n",
    "freq_all_hashtag"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('trump', 134),\n",
       " ('politics', 95),\n",
       " ('allahsoil', 92),\n",
       " ('libtard', 76),\n",
       " ('liberal', 75),\n",
       " ('sjw', 74),\n",
       " ('retweet', 63),\n",
       " ('miami', 46),\n",
       " ('black', 44),\n",
       " ('hate', 32)]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "freq_hatred_hashtag = FreqDist(list((\" \".join(df1[df1['label']==1]['hashtag'])).split())).most_common(10)\n",
    "freq_hatred_hashtag"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      words  frequency\n",
      "0      love       1591\n",
      "1  positive        879\n",
      "2     smile        577\n",
      "3   healthy        576\n",
      "4  thankful        471\n",
      "      words  frequency\n",
      "0      love       1591\n",
      "1  positive        879\n",
      "2     smile        577\n",
      "3   healthy        576\n",
      "4  thankful        471\n"
     ]
    }
   ],
   "source": [
    "df_allhashtag = pd.DataFrame(freq_all_hashtag, columns=['words', 'frequency'])\n",
    "df_hatredhashtag = pd.DataFrame(freq_hatred_hashtag, columns=['words', 'frequency'])\n",
    "print(df_allhashtag.head())\n",
    "print(df_allhashtag.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.barplot(x='words', y='frequency', data=df_allhashtag)\n",
    "plt.xticks(rotation = 45)\n",
    "plt.title('hashtag words frequency')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.barplot(x='words', y='frequency', data=df_hatredhashtag)\n",
    "plt.xticks(rotation = 45)\n",
    "plt.title('hatred hashtag frequency')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# to create sparse matrix corpus is created to pass to vectorizer\n",
    "len(df1)\n",
    "corpus = []\n",
    "for i in range(0,31962):\n",
    "    corpus.append(df1['normalized_text'][i])\n",
    "#corpus"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CountVectorizer(analyzer='word', binary=False, decode_error='strict',\n",
       "                dtype=<class 'numpy.int64'>, encoding='utf-8', input='content',\n",
       "                lowercase=True, max_df=1.0, max_features=None, min_df=1,\n",
       "                ngram_range=(1, 1), preprocessor=None,\n",
       "                stop_words=['i', 'me', 'my', 'myself', 'we', 'our', 'ours',\n",
       "                            'ourselves', 'you', \"you're\", \"you've\", \"you'll\",\n",
       "                            \"you'd\", 'your', 'yours', 'yourself', 'yourselves',\n",
       "                            'he', 'him', 'his', 'himself', 'she', \"she's\",\n",
       "                            'her', 'hers', 'herself', 'it', \"it's\", 'its',\n",
       "                            'itself', ...],\n",
       "                strip_accents=None, token_pattern='(?u)\\\\b\\\\w\\\\w+\\\\b',\n",
       "                tokenizer=None, vocabulary=None)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cv = CountVectorizer(stop_words=stopwords.words('english'))\n",
    "cv.fit(corpus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# creating dense matrix\n",
    "X = cv.transform(corpus).toarray()\n",
    "y = df1.iloc[:,1].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# train test split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Classification**\n",
    "\n",
    "**Logistic Regression (Linear Model)******"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=10, class_weight=None, dual=False, fit_intercept=True,\n",
       "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
       "                   multi_class='warn', n_jobs=None, penalty='l2',\n",
       "                   random_state=None, solver='warn', tol=0.0001, verbose=0,\n",
       "                   warm_start=False)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "classifier1 = LogisticRegression(C=10)\n",
    "classifier1.fit(X_train, y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = classifier1.predict(X_test)\n",
    "y_prob = classifier1.predict_proba(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.6702702702702703\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.97      0.99      0.98      9806\n",
      "           1       0.78      0.58      0.67       742\n",
      "\n",
      "    accuracy                           0.96     10548\n",
      "   macro avg       0.88      0.79      0.82     10548\n",
      "weighted avg       0.96      0.96      0.96     10548\n",
      "\n",
      "[[9687  119]\n",
      " [ 308  434]]\n"
     ]
    }
   ],
   "source": [
    "print(f1_score(y_test, y_pred))\n",
    "print(classification_report(y_test, y_pred))\n",
    "print(confusion_matrix(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**checking with TF-IDF vectorizer**\n",
    "*** Unigram, bi gram is used wih min_df = 10*******"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "tfidf = TfidfVectorizer(ngram_range=(1,3), min_df=10, stop_words=stopwords.words('english'))\n",
    "X1 = tfidf.fit_transform(corpus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=10, class_weight=None, dual=False, fit_intercept=True,\n",
       "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
       "                   multi_class='warn', n_jobs=None, penalty='l2',\n",
       "                   random_state=None, solver='warn', tol=0.0001, verbose=0,\n",
       "                   warm_start=False)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X1_train, X1_test, y1_train, y1_test = train_test_split(X1, y, test_size=0.33, random_state=42)\n",
    "classifier2 = LogisticRegression(C=10)\n",
    "classifier2.fit(X1_train, y1_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "y1_pred = classifier2.predict(X1_test)\n",
    "y1_prob = classifier2.predict_proba(X1_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.6308943089430895\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      0.99      0.98      9806\n",
      "           1       0.80      0.52      0.63       742\n",
      "\n",
      "    accuracy                           0.96     10548\n",
      "   macro avg       0.88      0.76      0.80     10548\n",
      "weighted avg       0.95      0.96      0.95     10548\n",
      "\n",
      "[[9706  100]\n",
      " [ 354  388]]\n"
     ]
    }
   ],
   "source": [
    "print(f1_score(y1_test, y1_pred))\n",
    "print(classification_report(y1_test, y1_pred))\n",
    "print(confusion_matrix(y1_test, y1_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "threshold = np.arange(0.1,0.9,0.1)\n",
    "score = [f1_score(y1_test, ((y1_prob[:,1] >= x).astype(int))) for x in threshold]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(threshold, score)\n",
    "plt.xlabel('Threshold Probability')\n",
    "plt.ylabel('F1 score')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**performing classification model on our test data** "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "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>id</th>\n",
       "      <th>tweet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>31963</td>\n",
       "      <td>#studiolife #aislife #requires #passion #dedic...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>31964</td>\n",
       "      <td>@user #white #supremacists want everyone to s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>31965</td>\n",
       "      <td>safe ways to heal your #acne!!    #altwaystohe...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>31966</td>\n",
       "      <td>is the hp and the cursed child book up for res...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>31967</td>\n",
       "      <td>3rd #bihday to my amazing, hilarious #nephew...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      id                                              tweet\n",
       "0  31963  #studiolife #aislife #requires #passion #dedic...\n",
       "1  31964   @user #white #supremacists want everyone to s...\n",
       "2  31965  safe ways to heal your #acne!!    #altwaystohe...\n",
       "3  31966  is the hp and the cursed child book up for res...\n",
       "4  31967    3rd #bihday to my amazing, hilarious #nephew..."
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2 = pd.read_csv('/kaggle/input/test.csv')\n",
    "df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2['normalized_text'] = df2['tweet'].apply(normalizer)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# creating corpus\n",
    "corpus_test = []\n",
    "for i in range(0,17197):\n",
    "    corpus_test.append(df2.normalized_text[i])\n",
    "#corpus_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "Test_X = tfidf.transform(corpus_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "pred_Y = classifier2.predict(Test_X)\n",
    "prob_Y = classifier2.predict_proba(Test_X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "df2['pred_label'] = pred_Y\n",
    "scores = (prob_Y[:,1] >= 0.5).astype(int)\n",
    "df2['score'] = scores\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "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>id</th>\n",
       "      <th>tweet</th>\n",
       "      <th>normalized_text</th>\n",
       "      <th>pred_label</th>\n",
       "      <th>score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>31982</td>\n",
       "      <td>thought factory: bbc neutrality on right wing ...</td>\n",
       "      <td>thought factory bbc neutrality right wing fasc...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>31989</td>\n",
       "      <td>chick gets fucked hottest naked lady</td>\n",
       "      <td>chick get fucked hottest naked lady</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>31993</td>\n",
       "      <td>i'll #never be #120 #again i'm   #i'm a #thick...</td>\n",
       "      <td>never thick woman blacktina</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>31996</td>\n",
       "      <td>suppo the #taiji fisherman! no bullying! no ra...</td>\n",
       "      <td>suppo taiji fisherman bullying racism tweet ta...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110</th>\n",
       "      <td>32073</td>\n",
       "      <td>hey @user - a $14000 ivanka bracelet? do you f...</td>\n",
       "      <td>hey ivanka bracelet feel good profiting xenoph...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>140</th>\n",
       "      <td>32103</td>\n",
       "      <td>@user all together this christmas: pls  &amp;amp; ...</td>\n",
       "      <td>together christmas pls amp follow national day...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>141</th>\n",
       "      <td>32104</td>\n",
       "      <td>you might be a libtard if... #libtard  #sjw #l...</td>\n",
       "      <td>might libtard libtard sjw liberal politics</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>159</th>\n",
       "      <td>32122</td>\n",
       "      <td>#people aren't protesting #trump because a #re...</td>\n",
       "      <td>people protesting trump republican trump fuher...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>160</th>\n",
       "      <td>32123</td>\n",
       "      <td>at work: attorneys for white officer who shot...</td>\n",
       "      <td>work attorney white officer shot philandocasti...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>164</th>\n",
       "      <td>32127</td>\n",
       "      <td>@user trump's long history of  explained 1970'...</td>\n",
       "      <td>trump long history explained thru nevermypresi...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>183</th>\n",
       "      <td>32146</td>\n",
       "      <td>@user  looked right at the cameraman as she gr...</td>\n",
       "      <td>looked right cameraman grabbed katie arm amp s...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>185</th>\n",
       "      <td>32148</td>\n",
       "      <td>looking up some african news out of boredem, s...</td>\n",
       "      <td>looking african news boredem stumbled upon oh ...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>261</th>\n",
       "      <td>32224</td>\n",
       "      <td>it could be worse. #embarrassed #unfounate #tr...</td>\n",
       "      <td>could worse embarrassed unfounate traumatized ...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>266</th>\n",
       "      <td>32229</td>\n",
       "      <td>@user @user it is still fucking bullshit that ...</td>\n",
       "      <td>still fucking bullshit giving bigot platform s...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>271</th>\n",
       "      <td>32234</td>\n",
       "      <td>@user @user @user it's always about race and r...</td>\n",
       "      <td>always race religion wrong w america</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>312</th>\n",
       "      <td>32275</td>\n",
       "      <td>#hate, , #crisis these are among many others t...</td>\n",
       "      <td>hate crisis among many others president elect ...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>351</th>\n",
       "      <td>32314</td>\n",
       "      <td>thought factory: left-right polarisation! #tru...</td>\n",
       "      <td>thought factory left right polarisation trump ...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>356</th>\n",
       "      <td>32319</td>\n",
       "      <td>@user go eat a shit dog you anti-american piec...</td>\n",
       "      <td>go eat shit dog anti american piece shit bet d...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>391</th>\n",
       "      <td>32354</td>\n",
       "      <td>have no idea who she is but what a stupid comm...</td>\n",
       "      <td>idea stupid comment make inviting response dea...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>403</th>\n",
       "      <td>32366</td>\n",
       "      <td>@user   you're  a typical ghetto pos politicia...</td>\n",
       "      <td>typical ghetto po politician refusing take pai...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>406</th>\n",
       "      <td>32369</td>\n",
       "      <td>the latest the knitting clio daily!  thanks to...</td>\n",
       "      <td>latest knitting clio daily thanks twitterstorians</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>422</th>\n",
       "      <td>32385</td>\n",
       "      <td>#easter! #meanwhile, at the #white #house......</td>\n",
       "      <td>easter meanwhile white house</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>427</th>\n",
       "      <td>32390</td>\n",
       "      <td>still no women. wow so .</td>\n",
       "      <td>still woman wow</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>429</th>\n",
       "      <td>32392</td>\n",
       "      <td>okey i know that the latest tweet was a bit cl...</td>\n",
       "      <td>okey know latest tweet bit clitchy hey like u ...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>430</th>\n",
       "      <td>32393</td>\n",
       "      <td>brutality rules new year on manus  close #manu...</td>\n",
       "      <td>brutality rule new year manus close manus ausp...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>493</th>\n",
       "      <td>32456</td>\n",
       "      <td>word factory: race denial prevents healing! #s...</td>\n",
       "      <td>word factory race denial prevents healing sout...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>498</th>\n",
       "      <td>32461</td>\n",
       "      <td>the latest the unsw daily!  thanks to @user @u...</td>\n",
       "      <td>latest unsw daily thanks ocean</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>541</th>\n",
       "      <td>32504</td>\n",
       "      <td>'@user  the #14th amendment  #blm discussed he...</td>\n",
       "      <td>user th amendment blm discussed thanks</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>590</th>\n",
       "      <td>32553</td>\n",
       "      <td>trump ally made  remarks towards obamas - cnn ...</td>\n",
       "      <td>trump ally made remark towards obamas cnn carl...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>604</th>\n",
       "      <td>32567</td>\n",
       "      <td>after trump, is the media normalizing the #kkk...</td>\n",
       "      <td>trump medium normalizing kkk alternet</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16697</th>\n",
       "      <td>48660</td>\n",
       "      <td>@user #jeffsessions  #attorneygeneral  from #a...</td>\n",
       "      <td>jeffsessions attorneygeneral alabama trump sta...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16702</th>\n",
       "      <td>48665</td>\n",
       "      <td>@user @user by \"many\", he means the 80% of eli...</td>\n",
       "      <td>many mean eligible u voter vote trump misogyny...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16706</th>\n",
       "      <td>48669</td>\n",
       "      <td>maybe. we need to teach our children and let t...</td>\n",
       "      <td>maybe need teach child let lead way police nee...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16727</th>\n",
       "      <td>48690</td>\n",
       "      <td>burning the american flag, really? where is th...</td>\n",
       "      <td>burning american flag really outrage unbelievable</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16765</th>\n",
       "      <td>48728</td>\n",
       "      <td>#christmas #trump #newyork co-chair makes  'go...</td>\n",
       "      <td>christmas trump newyork co chair make gorilla ...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16788</th>\n",
       "      <td>48751</td>\n",
       "      <td>@user @user keep reminding minority groups tha...</td>\n",
       "      <td>keep reminding minority group arent sma white ...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16793</th>\n",
       "      <td>48756</td>\n",
       "      <td>@user @user @user from a wonder of #creation, ...</td>\n",
       "      <td>wonder creation show cruelty n disgusting</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16800</th>\n",
       "      <td>48763</td>\n",
       "      <td>.@user or disabled journalist, rich jews, mexi...</td>\n",
       "      <td>user disabled journalist rich jew mexican rapi...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16805</th>\n",
       "      <td>48768</td>\n",
       "      <td>thought factory: bbc neutrality on right wing ...</td>\n",
       "      <td>thought factory bbc neutrality right wing fasc...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16814</th>\n",
       "      <td>48777</td>\n",
       "      <td>#ellenpompeo &amp;amp; the well-meaning-#white-#wo...</td>\n",
       "      <td>ellenpompeo amp well meaning white woman syndr...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16848</th>\n",
       "      <td>48811</td>\n",
       "      <td>@user @user @user you're a . 45% of cop killer...</td>\n",
       "      <td>cop killer black men population cop killer</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16866</th>\n",
       "      <td>48829</td>\n",
       "      <td>@user haha and he's saying that because he thi...</td>\n",
       "      <td>haha saying think male know female ponchobot</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16868</th>\n",
       "      <td>48831</td>\n",
       "      <td>@user @user the line \"don't be an idiot like m...</td>\n",
       "      <td>line idiot like human male line female would c...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16893</th>\n",
       "      <td>48856</td>\n",
       "      <td>the latest the blicqerâ¢ daily!  thanks to @u...</td>\n",
       "      <td>latest blicqer daily thanks maga</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16912</th>\n",
       "      <td>48875</td>\n",
       "      <td>i agree what he said was horrible but is this ...</td>\n",
       "      <td>agree said horrible really first time everyone...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16936</th>\n",
       "      <td>48899</td>\n",
       "      <td>@user ha! good riddance! #blacklivesmatter</td>\n",
       "      <td>ha good riddance blacklivesmatter</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16947</th>\n",
       "      <td>48910</td>\n",
       "      <td>#people aren't protesting #trump because a #re...</td>\n",
       "      <td>people protesting trump republican trump fuher...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16972</th>\n",
       "      <td>48935</td>\n",
       "      <td>â¦you #black &amp;amp; feel like the  are stompin...</td>\n",
       "      <td>black amp feel like stomping listen retweet ta...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16996</th>\n",
       "      <td>48959</td>\n",
       "      <td>@user @user @user #trumpusa when they call us ...</td>\n",
       "      <td>trumpusa call u homophobic misogynist really s...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16999</th>\n",
       "      <td>48962</td>\n",
       "      <td>office woman cutest girl on eah nude</td>\n",
       "      <td>office woman cutest girl eah nude</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17002</th>\n",
       "      <td>48965</td>\n",
       "      <td>@user you are a complete jerk. your comments a...</td>\n",
       "      <td>complete jerk comment reflection ignorance</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17031</th>\n",
       "      <td>48994</td>\n",
       "      <td>@user your comments are reflections of ignoran...</td>\n",
       "      <td>comment reflection ignorance</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17039</th>\n",
       "      <td>49002</td>\n",
       "      <td>@user #allahsoil we can correctly date the beg...</td>\n",
       "      <td>allahsoil correctly date beginning islamist id...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17044</th>\n",
       "      <td>49007</td>\n",
       "      <td>@user ok, i love ya, but stop the #bs. #obama ...</td>\n",
       "      <td>ok love ya stop b obama another bad newsrepo t...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17125</th>\n",
       "      <td>49088</td>\n",
       "      <td>be careful in criticizing #obama for his decis...</td>\n",
       "      <td>careful criticizing obama decision israel amp ...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17128</th>\n",
       "      <td>49091</td>\n",
       "      <td>@user the uk governmentâs new #anti-semitism...</td>\n",
       "      <td>uk government new anti semitism definition con...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17148</th>\n",
       "      <td>49111</td>\n",
       "      <td>we grew up fucked upð¤ its fucked upð¥ i'm...</td>\n",
       "      <td>grew fucked fucked believing better place fucking</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17176</th>\n",
       "      <td>49139</td>\n",
       "      <td>@user @user are the most racist pay ever!!!!!</td>\n",
       "      <td>racist pay ever</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17188</th>\n",
       "      <td>49151</td>\n",
       "      <td>black professor demonizes, proposes nazi style...</td>\n",
       "      <td>black professor demonizes proposes nazi style ...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17192</th>\n",
       "      <td>49155</td>\n",
       "      <td>thought factory: left-right polarisation! #tru...</td>\n",
       "      <td>thought factory left right polarisation trump ...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>814 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          id                                              tweet  \\\n",
       "19     31982  thought factory: bbc neutrality on right wing ...   \n",
       "26     31989              chick gets fucked hottest naked lady    \n",
       "30     31993  i'll #never be #120 #again i'm   #i'm a #thick...   \n",
       "33     31996  suppo the #taiji fisherman! no bullying! no ra...   \n",
       "110    32073  hey @user - a $14000 ivanka bracelet? do you f...   \n",
       "140    32103  @user all together this christmas: pls  &amp; ...   \n",
       "141    32104  you might be a libtard if... #libtard  #sjw #l...   \n",
       "159    32122  #people aren't protesting #trump because a #re...   \n",
       "160    32123   at work: attorneys for white officer who shot...   \n",
       "164    32127  @user trump's long history of  explained 1970'...   \n",
       "183    32146  @user  looked right at the cameraman as she gr...   \n",
       "185    32148  looking up some african news out of boredem, s...   \n",
       "261    32224  it could be worse. #embarrassed #unfounate #tr...   \n",
       "266    32229  @user @user it is still fucking bullshit that ...   \n",
       "271    32234  @user @user @user it's always about race and r...   \n",
       "312    32275  #hate, , #crisis these are among many others t...   \n",
       "351    32314  thought factory: left-right polarisation! #tru...   \n",
       "356    32319  @user go eat a shit dog you anti-american piec...   \n",
       "391    32354  have no idea who she is but what a stupid comm...   \n",
       "403    32366  @user   you're  a typical ghetto pos politicia...   \n",
       "406    32369  the latest the knitting clio daily!  thanks to...   \n",
       "422    32385    #easter! #meanwhile, at the #white #house......   \n",
       "427    32390                          still no women. wow so .    \n",
       "429    32392  okey i know that the latest tweet was a bit cl...   \n",
       "430    32393  brutality rules new year on manus  close #manu...   \n",
       "493    32456  word factory: race denial prevents healing! #s...   \n",
       "498    32461  the latest the unsw daily!  thanks to @user @u...   \n",
       "541    32504  '@user  the #14th amendment  #blm discussed he...   \n",
       "590    32553  trump ally made  remarks towards obamas - cnn ...   \n",
       "604    32567  after trump, is the media normalizing the #kkk...   \n",
       "...      ...                                                ...   \n",
       "16697  48660  @user #jeffsessions  #attorneygeneral  from #a...   \n",
       "16702  48665  @user @user by \"many\", he means the 80% of eli...   \n",
       "16706  48669  maybe. we need to teach our children and let t...   \n",
       "16727  48690  burning the american flag, really? where is th...   \n",
       "16765  48728  #christmas #trump #newyork co-chair makes  'go...   \n",
       "16788  48751  @user @user keep reminding minority groups tha...   \n",
       "16793  48756  @user @user @user from a wonder of #creation, ...   \n",
       "16800  48763  .@user or disabled journalist, rich jews, mexi...   \n",
       "16805  48768  thought factory: bbc neutrality on right wing ...   \n",
       "16814  48777  #ellenpompeo &amp; the well-meaning-#white-#wo...   \n",
       "16848  48811  @user @user @user you're a . 45% of cop killer...   \n",
       "16866  48829  @user haha and he's saying that because he thi...   \n",
       "16868  48831  @user @user the line \"don't be an idiot like m...   \n",
       "16893  48856  the latest the blicqerâ¢ daily!  thanks to @u...   \n",
       "16912  48875  i agree what he said was horrible but is this ...   \n",
       "16936  48899        @user ha! good riddance! #blacklivesmatter    \n",
       "16947  48910  #people aren't protesting #trump because a #re...   \n",
       "16972  48935  â¦you #black &amp; feel like the  are stompin...   \n",
       "16996  48959  @user @user @user #trumpusa when they call us ...   \n",
       "16999  48962              office woman cutest girl on eah nude    \n",
       "17002  48965  @user you are a complete jerk. your comments a...   \n",
       "17031  48994  @user your comments are reflections of ignoran...   \n",
       "17039  49002  @user #allahsoil we can correctly date the beg...   \n",
       "17044  49007  @user ok, i love ya, but stop the #bs. #obama ...   \n",
       "17125  49088  be careful in criticizing #obama for his decis...   \n",
       "17128  49091  @user the uk governmentâs new #anti-semitism...   \n",
       "17148  49111  we grew up fucked upð¤ its fucked upð¥ i'm...   \n",
       "17176  49139     @user @user are the most racist pay ever!!!!!    \n",
       "17188  49151  black professor demonizes, proposes nazi style...   \n",
       "17192  49155  thought factory: left-right polarisation! #tru...   \n",
       "\n",
       "                                         normalized_text  pred_label  score  \n",
       "19     thought factory bbc neutrality right wing fasc...           1      1  \n",
       "26                   chick get fucked hottest naked lady           1      1  \n",
       "30                           never thick woman blacktina           1      1  \n",
       "33     suppo taiji fisherman bullying racism tweet ta...           1      1  \n",
       "110    hey ivanka bracelet feel good profiting xenoph...           1      1  \n",
       "140    together christmas pls amp follow national day...           1      1  \n",
       "141           might libtard libtard sjw liberal politics           1      1  \n",
       "159    people protesting trump republican trump fuher...           1      1  \n",
       "160    work attorney white officer shot philandocasti...           1      1  \n",
       "164    trump long history explained thru nevermypresi...           1      1  \n",
       "183    looked right cameraman grabbed katie arm amp s...           1      1  \n",
       "185    looking african news boredem stumbled upon oh ...           1      1  \n",
       "261    could worse embarrassed unfounate traumatized ...           1      1  \n",
       "266    still fucking bullshit giving bigot platform s...           1      1  \n",
       "271                 always race religion wrong w america           1      1  \n",
       "312    hate crisis among many others president elect ...           1      1  \n",
       "351    thought factory left right polarisation trump ...           1      1  \n",
       "356    go eat shit dog anti american piece shit bet d...           1      1  \n",
       "391    idea stupid comment make inviting response dea...           1      1  \n",
       "403    typical ghetto po politician refusing take pai...           1      1  \n",
       "406    latest knitting clio daily thanks twitterstorians           1      1  \n",
       "422                         easter meanwhile white house           1      1  \n",
       "427                                      still woman wow           1      1  \n",
       "429    okey know latest tweet bit clitchy hey like u ...           1      1  \n",
       "430    brutality rule new year manus close manus ausp...           1      1  \n",
       "493    word factory race denial prevents healing sout...           1      1  \n",
       "498                       latest unsw daily thanks ocean           1      1  \n",
       "541               user th amendment blm discussed thanks           1      1  \n",
       "590    trump ally made remark towards obamas cnn carl...           1      1  \n",
       "604                trump medium normalizing kkk alternet           1      1  \n",
       "...                                                  ...         ...    ...  \n",
       "16697  jeffsessions attorneygeneral alabama trump sta...           1      1  \n",
       "16702  many mean eligible u voter vote trump misogyny...           1      1  \n",
       "16706  maybe need teach child let lead way police nee...           1      1  \n",
       "16727  burning american flag really outrage unbelievable           1      1  \n",
       "16765  christmas trump newyork co chair make gorilla ...           1      1  \n",
       "16788  keep reminding minority group arent sma white ...           1      1  \n",
       "16793          wonder creation show cruelty n disgusting           1      1  \n",
       "16800  user disabled journalist rich jew mexican rapi...           1      1  \n",
       "16805  thought factory bbc neutrality right wing fasc...           1      1  \n",
       "16814  ellenpompeo amp well meaning white woman syndr...           1      1  \n",
       "16848         cop killer black men population cop killer           1      1  \n",
       "16866       haha saying think male know female ponchobot           1      1  \n",
       "16868  line idiot like human male line female would c...           1      1  \n",
       "16893                   latest blicqer daily thanks maga           1      1  \n",
       "16912  agree said horrible really first time everyone...           1      1  \n",
       "16936                  ha good riddance blacklivesmatter           1      1  \n",
       "16947  people protesting trump republican trump fuher...           1      1  \n",
       "16972  black amp feel like stomping listen retweet ta...           1      1  \n",
       "16996  trumpusa call u homophobic misogynist really s...           1      1  \n",
       "16999                  office woman cutest girl eah nude           1      1  \n",
       "17002         complete jerk comment reflection ignorance           1      1  \n",
       "17031                       comment reflection ignorance           1      1  \n",
       "17039  allahsoil correctly date beginning islamist id...           1      1  \n",
       "17044  ok love ya stop b obama another bad newsrepo t...           1      1  \n",
       "17125  careful criticizing obama decision israel amp ...           1      1  \n",
       "17128  uk government new anti semitism definition con...           1      1  \n",
       "17148  grew fucked fucked believing better place fucking           1      1  \n",
       "17176                                    racist pay ever           1      1  \n",
       "17188  black professor demonizes proposes nazi style ...           1      1  \n",
       "17192  thought factory left right polarisation trump ...           1      1  \n",
       "\n",
       "[814 rows x 5 columns]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2[df2.pred_label == 1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}