Game Tweets Sentiment Analysis .ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import the python Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import re\n",
    "from textblob import TextBlob\n",
    "import nltk\n",
    "from wordcloud import WordCloud\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "plt.style.use('fivethirtyeight')\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read_file = pd.read_excel (r'Muzammil.xlsx')\n",
    "# read_file.to_csv (r'Gametweet.csv', index = None, header=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('Gametweet.csv', header = None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Ekaterina Vinokourova  the only daughter of Ru...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>eye do naught want two hear it  cyka blyat</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>Cryptic Cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>Cryptic Cyka that s a really good drawing man</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>(  )</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 39 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   0                                                  1   2   3   4   5   6   \\\n",
       "0   1  Ekaterina Vinokourova  the only daughter of Ru... NaN NaN NaN NaN NaN   \n",
       "1   1         eye do naught want two hear it  cyka blyat NaN NaN NaN NaN NaN   \n",
       "2   1   eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ... NaN NaN NaN NaN NaN   \n",
       "3   1                 Cryptic Cyka hmmmmmmmmm women hmmm NaN NaN NaN NaN NaN   \n",
       "4   1      Cryptic Cyka that s a really good drawing man NaN NaN NaN NaN NaN   \n",
       "\n",
       "   7   8     9   ...  29  30  31  32  33  34  35  36  37   38  \n",
       "0 NaN NaN   NaN  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "1 NaN NaN   NaN  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "2 NaN NaN   NaN  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "3 NaN NaN   NaN  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "4 NaN NaN  (  )  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "\n",
       "[5 rows x 39 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>290</th>\n",
       "      <td>0</td>\n",
       "      <td>I love thi game</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>291</th>\n",
       "      <td>0</td>\n",
       "      <td>game graphics is eye catching</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>292</th>\n",
       "      <td>0</td>\n",
       "      <td>Game level is really enjoyable</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>293</th>\n",
       "      <td>0</td>\n",
       "      <td>Good Pixel and very beautiful tool</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>294</th>\n",
       "      <td>0</td>\n",
       "      <td>you are free so come online and play with me</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 39 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     0                                              1   2   3   4   5   6   \\\n",
       "290   0                                I love thi game NaN NaN NaN NaN NaN   \n",
       "291   0                  game graphics is eye catching NaN NaN NaN NaN NaN   \n",
       "292   0                 Game level is really enjoyable NaN NaN NaN NaN NaN   \n",
       "293   0             Good Pixel and very beautiful tool NaN NaN NaN NaN NaN   \n",
       "294   0   you are free so come online and play with me NaN NaN NaN NaN NaN   \n",
       "\n",
       "     7   8    9   ...  29  30  31  32  33  34  35  36  37   38  \n",
       "290 NaN NaN  NaN  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "291 NaN NaN  NaN  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "292 NaN NaN  NaN  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "293 NaN NaN  NaN  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "294 NaN NaN  NaN  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "\n",
       "[5 rows x 39 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Data Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# rename the columns.\n",
    "df.rename(columns = {0:'target', 1 : 'tweets'}, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>target</th>\n",
       "      <th>tweets</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>29</th>\n",
       "      <th>30</th>\n",
       "      <th>31</th>\n",
       "      <th>32</th>\n",
       "      <th>33</th>\n",
       "      <th>34</th>\n",
       "      <th>35</th>\n",
       "      <th>36</th>\n",
       "      <th>37</th>\n",
       "      <th>38</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>Ekaterina Vinokourova  the only daughter of Ru...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>eye do naught want two hear it  cyka blyat</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>Cryptic Cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>Cryptic Cyka that s a really good drawing man</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>(  )</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 39 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   target                                             tweets   2   3   4   5  \\\n",
       "0       1  Ekaterina Vinokourova  the only daughter of Ru... NaN NaN NaN NaN   \n",
       "1       1         eye do naught want two hear it  cyka blyat NaN NaN NaN NaN   \n",
       "2       1   eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ... NaN NaN NaN NaN   \n",
       "3       1                 Cryptic Cyka hmmmmmmmmm women hmmm NaN NaN NaN NaN   \n",
       "4       1      Cryptic Cyka that s a really good drawing man NaN NaN NaN NaN   \n",
       "\n",
       "    6   7   8     9  ...  29  30  31  32  33  34  35  36  37   38  \n",
       "0 NaN NaN NaN   NaN  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "1 NaN NaN NaN   NaN  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "2 NaN NaN NaN   NaN  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "3 NaN NaN NaN   NaN  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "4 NaN NaN NaN  (  )  ... NaN NaN NaN NaN NaN NaN NaN NaN NaN  NaN  \n",
       "\n",
       "[5 rows x 39 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df[['tweets', 'target']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>tweets</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ekaterina Vinokourova  the only daughter of Ru...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>eye do naught want two hear it  cyka blyat</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cryptic Cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Cryptic Cyka that s a really good drawing man</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              tweets  target\n",
       "0  Ekaterina Vinokourova  the only daughter of Ru...       1\n",
       "1         eye do naught want two hear it  cyka blyat       1\n",
       "2   eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...       1\n",
       "3                 Cryptic Cyka hmmmmmmmmm women hmmm       1\n",
       "4      Cryptic Cyka that s a really good drawing man       1"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(295, 2)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 295 entries, 0 to 294\n",
      "Data columns (total 2 columns):\n",
      " #   Column  Non-Null Count  Dtype \n",
      "---  ------  --------------  ----- \n",
      " 0   tweets  295 non-null    object\n",
      " 1   target  295 non-null    int64 \n",
      "dtypes: int64(1), object(1)\n",
      "memory usage: 4.7+ KB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>295.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.111864</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.315735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           target\n",
       "count  295.000000\n",
       "mean     0.111864\n",
       "std      0.315735\n",
       "min      0.000000\n",
       "25%      0.000000\n",
       "50%      0.000000\n",
       "75%      0.000000\n",
       "max      1.000000"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['tweets', 'target'], dtype='object')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tweets    0\n",
       "target    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check missing values\n",
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check duplicate values\n",
    "df.duplicated().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(295, 2)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# remove duplicates value\n",
    "df = df.drop_duplicates()    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check duplicate values\n",
    "df.duplicated().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(289, 2)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tweets    0\n",
       "target    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2. Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>tweets</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ekaterina Vinokourova  the only daughter of Ru...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>eye do naught want two hear it  cyka blyat</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cryptic Cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Cryptic Cyka that s a really good drawing man</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              tweets  target\n",
       "0  Ekaterina Vinokourova  the only daughter of Ru...       1\n",
       "1         eye do naught want two hear it  cyka blyat       1\n",
       "2   eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...       1\n",
       "3                 Cryptic Cyka hmmmmmmmmm women hmmm       1\n",
       "4      Cryptic Cyka that s a really good drawing man       1"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    256\n",
       "1     33\n",
       "Name: target, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['target'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.pie(df['target'].value_counts(), labels = ['Positive', 'Toxic'], autopct = '%0.2f')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# calculate the number of characters \n",
    "df['num_of_characters'] = df['tweets'].apply(len)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "scrolled": false
   },
   "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>tweets</th>\n",
       "      <th>target</th>\n",
       "      <th>num_of_characters</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ekaterina Vinokourova  the only daughter of Ru...</td>\n",
       "      <td>1</td>\n",
       "      <td>303</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>eye do naught want two hear it  cyka blyat</td>\n",
       "      <td>1</td>\n",
       "      <td>42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...</td>\n",
       "      <td>1</td>\n",
       "      <td>221</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cryptic Cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Cryptic Cyka that s a really good drawing man</td>\n",
       "      <td>1</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              tweets  target  \\\n",
       "0  Ekaterina Vinokourova  the only daughter of Ru...       1   \n",
       "1         eye do naught want two hear it  cyka blyat       1   \n",
       "2   eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...       1   \n",
       "3                 Cryptic Cyka hmmmmmmmmm women hmmm       1   \n",
       "4      Cryptic Cyka that s a really good drawing man       1   \n",
       "\n",
       "   num_of_characters  \n",
       "0                303  \n",
       "1                 42  \n",
       "2                221  \n",
       "3                 35  \n",
       "4                 46  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# calculate number of words\n",
    "df['num_of_words'] = df['tweets'].apply(lambda x: len(nltk.word_tokenize(x)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>tweets</th>\n",
       "      <th>target</th>\n",
       "      <th>num_of_characters</th>\n",
       "      <th>num_of_words</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ekaterina Vinokourova  the only daughter of Ru...</td>\n",
       "      <td>1</td>\n",
       "      <td>303</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>eye do naught want two hear it  cyka blyat</td>\n",
       "      <td>1</td>\n",
       "      <td>42</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...</td>\n",
       "      <td>1</td>\n",
       "      <td>221</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cryptic Cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Cryptic Cyka that s a really good drawing man</td>\n",
       "      <td>1</td>\n",
       "      <td>46</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              tweets  target  \\\n",
       "0  Ekaterina Vinokourova  the only daughter of Ru...       1   \n",
       "1         eye do naught want two hear it  cyka blyat       1   \n",
       "2   eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...       1   \n",
       "3                 Cryptic Cyka hmmmmmmmmm women hmmm       1   \n",
       "4      Cryptic Cyka that s a really good drawing man       1   \n",
       "\n",
       "   num_of_characters  num_of_words  \n",
       "0                303            46  \n",
       "1                 42             9  \n",
       "2                221            41  \n",
       "3                 35             5  \n",
       "4                 46             9  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# calculate number of sentences\n",
    "df['num_of_sentences'] = df['tweets'].apply(lambda x: len(nltk.sent_tokenize(x)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>tweets</th>\n",
       "      <th>target</th>\n",
       "      <th>num_of_characters</th>\n",
       "      <th>num_of_words</th>\n",
       "      <th>num_of_sentences</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ekaterina Vinokourova  the only daughter of Ru...</td>\n",
       "      <td>1</td>\n",
       "      <td>303</td>\n",
       "      <td>46</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>eye do naught want two hear it  cyka blyat</td>\n",
       "      <td>1</td>\n",
       "      <td>42</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...</td>\n",
       "      <td>1</td>\n",
       "      <td>221</td>\n",
       "      <td>41</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cryptic Cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Cryptic Cyka that s a really good drawing man</td>\n",
       "      <td>1</td>\n",
       "      <td>46</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              tweets  target  \\\n",
       "0  Ekaterina Vinokourova  the only daughter of Ru...       1   \n",
       "1         eye do naught want two hear it  cyka blyat       1   \n",
       "2   eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...       1   \n",
       "3                 Cryptic Cyka hmmmmmmmmm women hmmm       1   \n",
       "4      Cryptic Cyka that s a really good drawing man       1   \n",
       "\n",
       "   num_of_characters  num_of_words  num_of_sentences  \n",
       "0                303            46                 1  \n",
       "1                 42             9                 1  \n",
       "2                221            41                 1  \n",
       "3                 35             5                 1  \n",
       "4                 46             9                 1  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "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>num_of_characters</th>\n",
       "      <th>num_of_words</th>\n",
       "      <th>num_of_sentences</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>289.000000</td>\n",
       "      <td>289.000000</td>\n",
       "      <td>289.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>129.927336</td>\n",
       "      <td>23.501730</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>80.070487</td>\n",
       "      <td>13.664157</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>16.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>62.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>116.000000</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>191.000000</td>\n",
       "      <td>33.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>351.000000</td>\n",
       "      <td>61.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       num_of_characters  num_of_words  num_of_sentences\n",
       "count         289.000000    289.000000             289.0\n",
       "mean          129.927336     23.501730               1.0\n",
       "std            80.070487     13.664157               0.0\n",
       "min            16.000000      4.000000               1.0\n",
       "25%            62.000000     12.000000               1.0\n",
       "50%           116.000000     22.000000               1.0\n",
       "75%           191.000000     33.000000               1.0\n",
       "max           351.000000     61.000000               1.0"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['num_of_characters', 'num_of_words', 'num_of_sentences']].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='num_of_characters', ylabel='Count'>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12,6))\n",
    "sns.histplot(df[df['target'] == 0]['num_of_characters'])\n",
    "sns.histplot(df[df['target'] == 1]['num_of_characters'], color='red')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='num_of_words', ylabel='Count'>"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12,6))\n",
    "sns.histplot(df[df['target'] == 0]['num_of_words'])\n",
    "sns.histplot(df[df['target'] == 1]['num_of_words'], color='red')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:xlabel='num_of_sentences', ylabel='Count'>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12,6))\n",
    "sns.histplot(df[df['target'] == 0]['num_of_sentences'])\n",
    "sns.histplot(df[df['target'] == 1]['num_of_sentences'], color='red')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Chacha Tech\\anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n",
      "C:\\Users\\Chacha Tech\\anaconda3\\lib\\site-packages\\seaborn\\distributions.py:305: UserWarning: Dataset has 0 variance; skipping density estimate.\n",
      "  warnings.warn(msg, UserWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x12f62927820>"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 582.375x540 with 12 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(df, hue='target')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3. Data Preprocessing\n",
    "## . Lower case\n",
    "## . Tokenization\n",
    "## . Removing special characters\n",
    "## . Removing stopwords and punctuations\n",
    "## . Stemming / Lemmitization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['i',\n",
       " 'me',\n",
       " 'my',\n",
       " 'myself',\n",
       " 'we',\n",
       " 'our',\n",
       " 'ours',\n",
       " 'ourselves',\n",
       " 'you',\n",
       " \"you're\",\n",
       " \"you've\",\n",
       " \"you'll\",\n",
       " \"you'd\",\n",
       " 'your',\n",
       " 'yours',\n",
       " 'yourself',\n",
       " 'yourselves',\n",
       " 'he',\n",
       " 'him',\n",
       " 'his',\n",
       " 'himself',\n",
       " 'she',\n",
       " \"she's\",\n",
       " 'her',\n",
       " 'hers',\n",
       " 'herself',\n",
       " 'it',\n",
       " \"it's\",\n",
       " 'its',\n",
       " 'itself',\n",
       " 'they',\n",
       " 'them',\n",
       " 'their',\n",
       " 'theirs',\n",
       " 'themselves',\n",
       " 'what',\n",
       " 'which',\n",
       " 'who',\n",
       " 'whom',\n",
       " 'this',\n",
       " 'that',\n",
       " \"that'll\",\n",
       " 'these',\n",
       " 'those',\n",
       " 'am',\n",
       " 'is',\n",
       " 'are',\n",
       " 'was',\n",
       " 'were',\n",
       " 'be',\n",
       " 'been',\n",
       " 'being',\n",
       " 'have',\n",
       " 'has',\n",
       " 'had',\n",
       " 'having',\n",
       " 'do',\n",
       " 'does',\n",
       " 'did',\n",
       " 'doing',\n",
       " 'a',\n",
       " 'an',\n",
       " 'the',\n",
       " 'and',\n",
       " 'but',\n",
       " 'if',\n",
       " 'or',\n",
       " 'because',\n",
       " 'as',\n",
       " 'until',\n",
       " 'while',\n",
       " 'of',\n",
       " 'at',\n",
       " 'by',\n",
       " 'for',\n",
       " 'with',\n",
       " 'about',\n",
       " 'against',\n",
       " 'between',\n",
       " 'into',\n",
       " 'through',\n",
       " 'during',\n",
       " 'before',\n",
       " 'after',\n",
       " 'above',\n",
       " 'below',\n",
       " 'to',\n",
       " 'from',\n",
       " 'up',\n",
       " 'down',\n",
       " 'in',\n",
       " 'out',\n",
       " 'on',\n",
       " 'off',\n",
       " 'over',\n",
       " 'under',\n",
       " 'again',\n",
       " 'further',\n",
       " 'then',\n",
       " 'once',\n",
       " 'here',\n",
       " 'there',\n",
       " 'when',\n",
       " 'where',\n",
       " 'why',\n",
       " 'how',\n",
       " 'all',\n",
       " 'any',\n",
       " 'both',\n",
       " 'each',\n",
       " 'few',\n",
       " 'more',\n",
       " 'most',\n",
       " 'other',\n",
       " 'some',\n",
       " 'such',\n",
       " 'no',\n",
       " 'nor',\n",
       " 'not',\n",
       " 'only',\n",
       " 'own',\n",
       " 'same',\n",
       " 'so',\n",
       " 'than',\n",
       " 'too',\n",
       " 'very',\n",
       " 's',\n",
       " 't',\n",
       " 'can',\n",
       " 'will',\n",
       " 'just',\n",
       " 'don',\n",
       " \"don't\",\n",
       " 'should',\n",
       " \"should've\",\n",
       " 'now',\n",
       " 'd',\n",
       " 'll',\n",
       " 'm',\n",
       " 'o',\n",
       " 're',\n",
       " 've',\n",
       " 'y',\n",
       " 'ain',\n",
       " 'aren',\n",
       " \"aren't\",\n",
       " 'couldn',\n",
       " \"couldn't\",\n",
       " 'didn',\n",
       " \"didn't\",\n",
       " 'doesn',\n",
       " \"doesn't\",\n",
       " 'hadn',\n",
       " \"hadn't\",\n",
       " 'hasn',\n",
       " \"hasn't\",\n",
       " 'haven',\n",
       " \"haven't\",\n",
       " 'isn',\n",
       " \"isn't\",\n",
       " 'ma',\n",
       " 'mightn',\n",
       " \"mightn't\",\n",
       " 'mustn',\n",
       " \"mustn't\",\n",
       " 'needn',\n",
       " \"needn't\",\n",
       " 'shan',\n",
       " \"shan't\",\n",
       " 'shouldn',\n",
       " \"shouldn't\",\n",
       " 'wasn',\n",
       " \"wasn't\",\n",
       " 'weren',\n",
       " \"weren't\",\n",
       " 'won',\n",
       " \"won't\",\n",
       " 'wouldn',\n",
       " \"wouldn't\"]"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from nltk.corpus import stopwords\n",
    "stopwords.words('english')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'!\"#$%&\\'()*+,-./:;<=>?@[\\\\]^_`{|}~'"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import string\n",
    "string.punctuation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'love'"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from nltk.stem.porter import PorterStemmer\n",
    "ps = PorterStemmer()\n",
    "ps.stem('loving')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "def transform_text(text):\n",
    "    text = text.lower()                  # 1. lower case\n",
    "    text = nltk.word_tokenize(text)      # 2. Tokenize\n",
    "   \n",
    "    y = []\n",
    "    for i in text:\n",
    "        if i.isalnum():                  # 3. remove special characters\n",
    "            y.append(i)\n",
    "            \n",
    "    text = y[:]\n",
    "    y.clear()\n",
    "    \n",
    "    for i in text:\n",
    "        if i not in stopwords.words('english') and i not in string.punctuation: #  4. stopwords and punctuation\n",
    "            y.append(i)\n",
    "            \n",
    "    text = y[:]\n",
    "    y.clear()\n",
    "    \n",
    "    for i in text:\n",
    "        y.append(ps.stem(i)) # 5. Stemming\n",
    "    \n",
    "    \n",
    "    return \" \".join(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'hi'"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transform_text('Hi How are you') # 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'hi'"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transform_text('Hi How are you') # 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'hi'"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transform_text('Hi How are %% you') # 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'like present data scienc'"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transform_text('Did you like my presentation on Data Science?') # 4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'love machin learn lectur youtub channel'"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "transform_text('I loved the machine learning lectures on youtube channel. How about you?') # 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "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>tweets</th>\n",
       "      <th>target</th>\n",
       "      <th>num_of_characters</th>\n",
       "      <th>num_of_words</th>\n",
       "      <th>num_of_sentences</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ekaterina Vinokourova  the only daughter of Ru...</td>\n",
       "      <td>1</td>\n",
       "      <td>303</td>\n",
       "      <td>46</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>eye do naught want two hear it  cyka blyat</td>\n",
       "      <td>1</td>\n",
       "      <td>42</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...</td>\n",
       "      <td>1</td>\n",
       "      <td>221</td>\n",
       "      <td>41</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cryptic Cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Cryptic Cyka that s a really good drawing man</td>\n",
       "      <td>1</td>\n",
       "      <td>46</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              tweets  target  \\\n",
       "0  Ekaterina Vinokourova  the only daughter of Ru...       1   \n",
       "1         eye do naught want two hear it  cyka blyat       1   \n",
       "2   eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...       1   \n",
       "3                 Cryptic Cyka hmmmmmmmmm women hmmm       1   \n",
       "4      Cryptic Cyka that s a really good drawing man       1   \n",
       "\n",
       "   num_of_characters  num_of_words  num_of_sentences  \n",
       "0                303            46                 1  \n",
       "1                 42             9                 1  \n",
       "2                221            41                 1  \n",
       "3                 35             5                 1  \n",
       "4                 46             9                 1  "
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['transformed_tweets'] = df['tweets'].apply(transform_text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "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>tweets</th>\n",
       "      <th>target</th>\n",
       "      <th>num_of_characters</th>\n",
       "      <th>num_of_words</th>\n",
       "      <th>num_of_sentences</th>\n",
       "      <th>transformed_tweets</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ekaterina Vinokourova  the only daughter of Ru...</td>\n",
       "      <td>1</td>\n",
       "      <td>303</td>\n",
       "      <td>46</td>\n",
       "      <td>1</td>\n",
       "      <td>ekaterina vinokourova daughter russian foreign...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>eye do naught want two hear it  cyka blyat</td>\n",
       "      <td>1</td>\n",
       "      <td>42</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>eye naught want two hear cyka blyat</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...</td>\n",
       "      <td>1</td>\n",
       "      <td>221</td>\n",
       "      <td>41</td>\n",
       "      <td>1</td>\n",
       "      <td>ew fkgam gs xw rwapodcast cyka bro supper offe...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cryptic Cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>cryptic cyka hmmmmmmmmm women hmmm</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Cryptic Cyka that s a really good drawing man</td>\n",
       "      <td>1</td>\n",
       "      <td>46</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>cryptic cyka realli good draw man</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              tweets  target  \\\n",
       "0  Ekaterina Vinokourova  the only daughter of Ru...       1   \n",
       "1         eye do naught want two hear it  cyka blyat       1   \n",
       "2   eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...       1   \n",
       "3                 Cryptic Cyka hmmmmmmmmm women hmmm       1   \n",
       "4      Cryptic Cyka that s a really good drawing man       1   \n",
       "\n",
       "   num_of_characters  num_of_words  num_of_sentences  \\\n",
       "0                303            46                 1   \n",
       "1                 42             9                 1   \n",
       "2                221            41                 1   \n",
       "3                 35             5                 1   \n",
       "4                 46             9                 1   \n",
       "\n",
       "                                  transformed_tweets  \n",
       "0  ekaterina vinokourova daughter russian foreign...  \n",
       "1                eye naught want two hear cyka blyat  \n",
       "2  ew fkgam gs xw rwapodcast cyka bro supper offe...  \n",
       "3                 cryptic cyka hmmmmmmmmm women hmmm  \n",
       "4                  cryptic cyka realli good draw man  "
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     ekaterina vinokourova daughter russian foreign...\n",
       "1                   eye naught want two hear cyka blyat\n",
       "2     ew fkgam gs xw rwapodcast cyka bro supper offe...\n",
       "3                    cryptic cyka hmmmmmmmmm women hmmm\n",
       "4                     cryptic cyka realli good draw man\n",
       "5                                cryptic cyka know deal\n",
       "6     teacher didnt wipe sinc friday fridaynightfunk...\n",
       "7                                 aregularnam wait mean\n",
       "8                        thing id let http co gnzyv tcq\n",
       "9                                   cryptic cyka god lt\n",
       "10                                    yepyep cyka blyat\n",
       "11    nigellovesy soooo overr fuck noob game follow ...\n",
       "12    namastekikii liter mean bitch idiomat use stan...\n",
       "13                   cykaart niceee congrat cyka deserv\n",
       "14    tagaan kog chanc mag learn ug new languag pick...\n",
       "15    defenceu sad pathet putin need help china merc...\n",
       "16                       taliaa novaa cyka shut fuck nm\n",
       "17    mfa russia mfa ukrain rusembukrain osc rf osc ...\n",
       "18    un american lmao say fuck shit ass cyka blyat ...\n",
       "19    thank yuri teach russian phrase today learnt c...\n",
       "20                                           cyka blyat\n",
       "21             know russian cyka blyat http co tex mqmt\n",
       "22    marketrebel sooner later game server gon na fu...\n",
       "23    cyka blyattttt katiewonn hi friend name maxwar...\n",
       "24    rogozin profil pic budget photoshop caus head ...\n",
       "25    brokenboi get absolut fuck own stupid russian ...\n",
       "26                           barackobama eat cyka blyat\n",
       "27            ubercsgo timjdillon kiev fight cyka blyat\n",
       "28                       gg fuck shit game cyka toosefn\n",
       "29    elonmusk challeng cs go match could shout cyka...\n",
       "30                               cyka say cyka say like\n",
       "31    eigenrobot true even heartbreak ig gen z russi...\n",
       "32            tsybulskaliubov cyka blyat give sugar die\n",
       "Name: transformed_tweets, dtype: object"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df['target'] == 1]['transformed_tweets']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "33        144 therealmurban lunaoi vn anoth whitoid lmao\n",
       "34     145 marku redmenac lunaoi vn funni fanbas priv...\n",
       "35     146 srf j leav xtreme game http co ytzdli ii r...\n",
       "36     147 srf j leav xtreme game replac old eleven l...\n",
       "37     148 dpc cn pwrd dota underway betnow http co b...\n",
       "                             ...                        \n",
       "290                                        love thi game\n",
       "291                               game graphic eye catch\n",
       "292                              game level realli enjoy\n",
       "293                               good pixel beauti tool\n",
       "294                                 free come onlin play\n",
       "Name: transformed_tweets, Length: 256, dtype: object"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df['target'] == 0]['transformed_tweets']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create a function to get subjectivity.\n",
    "def getSubjectivity(tweet):\n",
    "    return TextBlob(tweet).sentiment.subjectivity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create a function to get polarity.\n",
    "def getPolarity(tweet):\n",
    "    return TextBlob(tweet).sentiment.polarity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sentiment(polarity=-0.5, subjectivity=1.0)\n",
      "Sentiment(polarity=0.5, subjectivity=0.6)\n"
     ]
    }
   ],
   "source": [
    "f1 = \"Not awesome\"\n",
    "f2 = \"I love this food .Awesome\"\n",
    "\n",
    "blob1 = TextBlob(f1) \n",
    "blob2 = TextBlob(f2) \n",
    "\n",
    "print(blob1.sentiment)\n",
    "print(blob2.sentiment)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # create two new columns.\n",
    "df['subjectivity'] = df['transformed_tweets'].apply(getSubjectivity)\n",
    "df['polarity'] = df['transformed_tweets'].apply(getPolarity)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "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>tweets</th>\n",
       "      <th>target</th>\n",
       "      <th>num_of_characters</th>\n",
       "      <th>num_of_words</th>\n",
       "      <th>num_of_sentences</th>\n",
       "      <th>transformed_tweets</th>\n",
       "      <th>subjectivity</th>\n",
       "      <th>polarity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ekaterina Vinokourova  the only daughter of Ru...</td>\n",
       "      <td>1</td>\n",
       "      <td>303</td>\n",
       "      <td>46</td>\n",
       "      <td>1</td>\n",
       "      <td>ekaterina vinokourova daughter russian foreign...</td>\n",
       "      <td>0.056250</td>\n",
       "      <td>-0.05625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>eye do naught want two hear it  cyka blyat</td>\n",
       "      <td>1</td>\n",
       "      <td>42</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>eye naught want two hear cyka blyat</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...</td>\n",
       "      <td>1</td>\n",
       "      <td>221</td>\n",
       "      <td>41</td>\n",
       "      <td>1</td>\n",
       "      <td>ew fkgam gs xw rwapodcast cyka bro supper offe...</td>\n",
       "      <td>0.523254</td>\n",
       "      <td>0.16881</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cryptic Cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>cryptic cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Cryptic Cyka that s a really good drawing man</td>\n",
       "      <td>1</td>\n",
       "      <td>46</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>cryptic cyka realli good draw man</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.70000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              tweets  target  \\\n",
       "0  Ekaterina Vinokourova  the only daughter of Ru...       1   \n",
       "1         eye do naught want two hear it  cyka blyat       1   \n",
       "2   eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...       1   \n",
       "3                 Cryptic Cyka hmmmmmmmmm women hmmm       1   \n",
       "4      Cryptic Cyka that s a really good drawing man       1   \n",
       "\n",
       "   num_of_characters  num_of_words  num_of_sentences  \\\n",
       "0                303            46                 1   \n",
       "1                 42             9                 1   \n",
       "2                221            41                 1   \n",
       "3                 35             5                 1   \n",
       "4                 46             9                 1   \n",
       "\n",
       "                                  transformed_tweets  subjectivity  polarity  \n",
       "0  ekaterina vinokourova daughter russian foreign...      0.056250  -0.05625  \n",
       "1                eye naught want two hear cyka blyat      0.000000   0.00000  \n",
       "2  ew fkgam gs xw rwapodcast cyka bro supper offe...      0.523254   0.16881  \n",
       "3                 cryptic cyka hmmmmmmmmm women hmmm      0.000000   0.00000  \n",
       "4                  cryptic cyka realli good draw man      0.600000   0.70000  "
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create a function to compute the negative , neutral and positive tweet analysis.\n",
    "def getAnalysis(score):\n",
    "    if score < 0:\n",
    "        return 'negative'\n",
    "    elif score == 0:\n",
    "        return 'neutral'\n",
    "    else:\n",
    "        return 'positive'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "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>tweets</th>\n",
       "      <th>target</th>\n",
       "      <th>num_of_characters</th>\n",
       "      <th>num_of_words</th>\n",
       "      <th>num_of_sentences</th>\n",
       "      <th>transformed_tweets</th>\n",
       "      <th>subjectivity</th>\n",
       "      <th>polarity</th>\n",
       "      <th>analysis</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ekaterina Vinokourova  the only daughter of Ru...</td>\n",
       "      <td>1</td>\n",
       "      <td>303</td>\n",
       "      <td>46</td>\n",
       "      <td>1</td>\n",
       "      <td>ekaterina vinokourova daughter russian foreign...</td>\n",
       "      <td>0.056250</td>\n",
       "      <td>-0.05625</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>eye do naught want two hear it  cyka blyat</td>\n",
       "      <td>1</td>\n",
       "      <td>42</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>eye naught want two hear cyka blyat</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...</td>\n",
       "      <td>1</td>\n",
       "      <td>221</td>\n",
       "      <td>41</td>\n",
       "      <td>1</td>\n",
       "      <td>ew fkgam gs xw rwapodcast cyka bro supper offe...</td>\n",
       "      <td>0.523254</td>\n",
       "      <td>0.16881</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cryptic Cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>cryptic cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Cryptic Cyka that s a really good drawing man</td>\n",
       "      <td>1</td>\n",
       "      <td>46</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>cryptic cyka realli good draw man</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.70000</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Cryptic Cyka they knows her deal</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>cryptic cyka know deal</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Teacher didnt wipe her off since friday  frida...</td>\n",
       "      <td>1</td>\n",
       "      <td>82</td>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "      <td>teacher didnt wipe sinc friday fridaynightfunk...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Aregularname Wait what does that mean</td>\n",
       "      <td>1</td>\n",
       "      <td>39</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>aregularnam wait mean</td>\n",
       "      <td>0.687500</td>\n",
       "      <td>-0.31250</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>The things id let him do   https   t co gnZYv ...</td>\n",
       "      <td>1</td>\n",
       "      <td>50</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>thing id let http co gnzyv tcq</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Cryptic Cyka on god  lt</td>\n",
       "      <td>1</td>\n",
       "      <td>26</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>cryptic cyka god lt</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              tweets  target  \\\n",
       "0  Ekaterina Vinokourova  the only daughter of Ru...       1   \n",
       "1         eye do naught want two hear it  cyka blyat       1   \n",
       "2   eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...       1   \n",
       "3                 Cryptic Cyka hmmmmmmmmm women hmmm       1   \n",
       "4      Cryptic Cyka that s a really good drawing man       1   \n",
       "5                   Cryptic Cyka they knows her deal       1   \n",
       "6  Teacher didnt wipe her off since friday  frida...       1   \n",
       "7             Aregularname Wait what does that mean        1   \n",
       "8  The things id let him do   https   t co gnZYv ...       1   \n",
       "9                          Cryptic Cyka on god  lt         1   \n",
       "\n",
       "   num_of_characters  num_of_words  num_of_sentences  \\\n",
       "0                303            46                 1   \n",
       "1                 42             9                 1   \n",
       "2                221            41                 1   \n",
       "3                 35             5                 1   \n",
       "4                 46             9                 1   \n",
       "5                 33             6                 1   \n",
       "6                 82            12                 1   \n",
       "7                 39             6                 1   \n",
       "8                 50            11                 1   \n",
       "9                 26             5                 1   \n",
       "\n",
       "                                  transformed_tweets  subjectivity  polarity  \\\n",
       "0  ekaterina vinokourova daughter russian foreign...      0.056250  -0.05625   \n",
       "1                eye naught want two hear cyka blyat      0.000000   0.00000   \n",
       "2  ew fkgam gs xw rwapodcast cyka bro supper offe...      0.523254   0.16881   \n",
       "3                 cryptic cyka hmmmmmmmmm women hmmm      0.000000   0.00000   \n",
       "4                  cryptic cyka realli good draw man      0.600000   0.70000   \n",
       "5                             cryptic cyka know deal      0.000000   0.00000   \n",
       "6  teacher didnt wipe sinc friday fridaynightfunk...      0.000000   0.00000   \n",
       "7                              aregularnam wait mean      0.687500  -0.31250   \n",
       "8                     thing id let http co gnzyv tcq      0.000000   0.00000   \n",
       "9                                cryptic cyka god lt      0.000000   0.00000   \n",
       "\n",
       "   analysis  \n",
       "0  negative  \n",
       "1   neutral  \n",
       "2  positive  \n",
       "3   neutral  \n",
       "4  positive  \n",
       "5   neutral  \n",
       "6   neutral  \n",
       "7  negative  \n",
       "8   neutral  \n",
       "9   neutral  "
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create a new column for tweet analysis.\n",
    "df['analysis'] = df['polarity'].apply(getAnalysis)\n",
    "df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "spam_corpus = []\n",
    "for msg in df[df['target'] == 1]['transformed_tweets'].tolist():\n",
    "    for word in msg.split():\n",
    "        spam_corpus.append(word)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['ekaterina',\n",
       " 'vinokourova',\n",
       " 'daughter',\n",
       " 'russian',\n",
       " 'foreign',\n",
       " 'minist',\n",
       " 'sergei',\n",
       " 'lavrov',\n",
       " 'spent',\n",
       " 'year',\n",
       " 'u',\n",
       " 'studi',\n",
       " 'polit',\n",
       " 'scienc',\n",
       " 'columbia',\n",
       " 'univers',\n",
       " 'econom',\n",
       " 'london',\n",
       " 'school',\n",
       " 'econom',\n",
       " 'cyka',\n",
       " 'go',\n",
       " 'go',\n",
       " 'back',\n",
       " 'motherfuckingruissia',\n",
       " 'http',\n",
       " 'co',\n",
       " 'znkcrdqazz',\n",
       " 'eye',\n",
       " 'naught',\n",
       " 'want',\n",
       " 'two',\n",
       " 'hear',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'ew',\n",
       " 'fkgam',\n",
       " 'gs',\n",
       " 'xw',\n",
       " 'rwapodcast',\n",
       " 'cyka',\n",
       " 'bro',\n",
       " 'supper',\n",
       " 'offens',\n",
       " 'tell',\n",
       " 'also',\n",
       " 'call',\n",
       " 'bori',\n",
       " 'channel',\n",
       " 'ye',\n",
       " 'dab',\n",
       " 'hard',\n",
       " 'russian',\n",
       " 'bro',\n",
       " 'yeah',\n",
       " 'sure',\n",
       " 'also',\n",
       " 'dub',\n",
       " 'right',\n",
       " 'true',\n",
       " 'mememan',\n",
       " 'cryptic',\n",
       " 'cyka',\n",
       " 'hmmmmmmmmm',\n",
       " 'women',\n",
       " 'hmmm',\n",
       " 'cryptic',\n",
       " 'cyka',\n",
       " 'realli',\n",
       " 'good',\n",
       " 'draw',\n",
       " 'man',\n",
       " 'cryptic',\n",
       " 'cyka',\n",
       " 'know',\n",
       " 'deal',\n",
       " 'teacher',\n",
       " 'didnt',\n",
       " 'wipe',\n",
       " 'sinc',\n",
       " 'friday',\n",
       " 'fridaynightfunkin',\n",
       " 'http',\n",
       " 'co',\n",
       " 'eorhjtxhsi',\n",
       " 'aregularnam',\n",
       " 'wait',\n",
       " 'mean',\n",
       " 'thing',\n",
       " 'id',\n",
       " 'let',\n",
       " 'http',\n",
       " 'co',\n",
       " 'gnzyv',\n",
       " 'tcq',\n",
       " 'cryptic',\n",
       " 'cyka',\n",
       " 'god',\n",
       " 'lt',\n",
       " 'yepyep',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'nigellovesy',\n",
       " 'soooo',\n",
       " 'overr',\n",
       " 'fuck',\n",
       " 'noob',\n",
       " 'game',\n",
       " 'follow',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'namastekikii',\n",
       " 'liter',\n",
       " 'mean',\n",
       " 'bitch',\n",
       " 'idiomat',\n",
       " 'use',\n",
       " 'stand',\n",
       " 'sort',\n",
       " 'explet',\n",
       " 'use',\n",
       " 'english',\n",
       " 'googl',\n",
       " 'cyka',\n",
       " 'interest',\n",
       " 'cykaart',\n",
       " 'niceee',\n",
       " 'congrat',\n",
       " 'cyka',\n",
       " 'deserv',\n",
       " 'tagaan',\n",
       " 'kog',\n",
       " 'chanc',\n",
       " 'mag',\n",
       " 'learn',\n",
       " 'ug',\n",
       " 'new',\n",
       " 'languag',\n",
       " 'pick',\n",
       " 'german',\n",
       " 'russian',\n",
       " 'cyka',\n",
       " 'bahug',\n",
       " 'bylat',\n",
       " 'defenceu',\n",
       " 'sad',\n",
       " 'pathet',\n",
       " 'putin',\n",
       " 'need',\n",
       " 'help',\n",
       " 'china',\n",
       " 'mercenari',\n",
       " 'syria',\n",
       " 'attack',\n",
       " 'ukrain',\n",
       " 'littl',\n",
       " 'cyka',\n",
       " 'glori',\n",
       " 'ukrain',\n",
       " 'taliaa',\n",
       " 'novaa',\n",
       " 'cyka',\n",
       " 'shut',\n",
       " 'fuck',\n",
       " 'nm',\n",
       " 'mfa',\n",
       " 'russia',\n",
       " 'mfa',\n",
       " 'ukrain',\n",
       " 'rusembukrain',\n",
       " 'osc',\n",
       " 'rf',\n",
       " 'osc',\n",
       " 'russiaun',\n",
       " 'dpol',\n",
       " 'un',\n",
       " 'rusembusa',\n",
       " 'look',\n",
       " 'like',\n",
       " 'gentli',\n",
       " 'place',\n",
       " 'damag',\n",
       " 'ground',\n",
       " 'surround',\n",
       " 'area',\n",
       " 'cyka',\n",
       " 'nobodi',\n",
       " 'believ',\n",
       " 'un',\n",
       " 'american',\n",
       " 'lmao',\n",
       " 'say',\n",
       " 'fuck',\n",
       " 'shit',\n",
       " 'ass',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'bird',\n",
       " 'app',\n",
       " 'time',\n",
       " 'moreso',\n",
       " 'dont',\n",
       " 'trust',\n",
       " 'peopl',\n",
       " 'whose',\n",
       " 'first',\n",
       " 'instinct',\n",
       " 'see',\n",
       " 'someth',\n",
       " 'dislik',\n",
       " 'rage',\n",
       " 'work',\n",
       " 'class',\n",
       " 'need',\n",
       " 'leadership',\n",
       " 'take',\n",
       " 'sec',\n",
       " 'paus',\n",
       " 'think',\n",
       " 'move',\n",
       " 'advanc',\n",
       " 'thank',\n",
       " 'yuri',\n",
       " 'teach',\n",
       " 'russian',\n",
       " 'phrase',\n",
       " 'today',\n",
       " 'learnt',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'mean',\n",
       " 'friendship',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'know',\n",
       " 'russian',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'http',\n",
       " 'co',\n",
       " 'tex',\n",
       " 'mqmt',\n",
       " 'marketrebel',\n",
       " 'sooner',\n",
       " 'later',\n",
       " 'game',\n",
       " 'server',\n",
       " 'gon',\n",
       " 'na',\n",
       " 'full',\n",
       " 'kid',\n",
       " 'scream',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'cyka',\n",
       " 'blyattttt',\n",
       " 'katiewonn',\n",
       " 'hi',\n",
       " 'friend',\n",
       " 'name',\n",
       " 'maxward',\n",
       " 'year',\n",
       " 'realli',\n",
       " 'need',\n",
       " 'help',\n",
       " 'pleas',\n",
       " 'ignor',\n",
       " 'messag',\n",
       " 'pleas',\n",
       " 'enough',\n",
       " 'whatsapp',\n",
       " 'number',\n",
       " 'thank',\n",
       " 'rogozin',\n",
       " 'profil',\n",
       " 'pic',\n",
       " 'budget',\n",
       " 'photoshop',\n",
       " 'caus',\n",
       " 'head',\n",
       " 'way',\n",
       " 'big',\n",
       " 'fat',\n",
       " 'ass',\n",
       " 'photo',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'fake',\n",
       " 'brokenboi',\n",
       " 'get',\n",
       " 'absolut',\n",
       " 'fuck',\n",
       " 'own',\n",
       " 'stupid',\n",
       " 'russian',\n",
       " 'putin',\n",
       " 'support',\n",
       " 'nazi',\n",
       " 'cunt',\n",
       " 'post',\n",
       " 'peopl',\n",
       " 'block',\n",
       " 'look',\n",
       " 'decid',\n",
       " 'deactiv',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'fuck',\n",
       " 'melon',\n",
       " 'http',\n",
       " 'co',\n",
       " 'f',\n",
       " 'c',\n",
       " 'hywsyu',\n",
       " 'barackobama',\n",
       " 'eat',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'ubercsgo',\n",
       " 'timjdillon',\n",
       " 'kiev',\n",
       " 'fight',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'gg',\n",
       " 'fuck',\n",
       " 'shit',\n",
       " 'game',\n",
       " 'cyka',\n",
       " 'toosefn',\n",
       " 'elonmusk',\n",
       " 'challeng',\n",
       " 'cs',\n",
       " 'go',\n",
       " 'match',\n",
       " 'could',\n",
       " 'shout',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'teabag',\n",
       " 'corps',\n",
       " 'cyka',\n",
       " 'say',\n",
       " 'cyka',\n",
       " 'say',\n",
       " 'like',\n",
       " 'eigenrobot',\n",
       " 'true',\n",
       " 'even',\n",
       " 'heartbreak',\n",
       " 'ig',\n",
       " 'gen',\n",
       " 'z',\n",
       " 'russian',\n",
       " 'front',\n",
       " 'line',\n",
       " 'go',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'line',\n",
       " 'surli',\n",
       " 'mercenari',\n",
       " 'prod',\n",
       " 'forward',\n",
       " 'els',\n",
       " 'make',\n",
       " 'even',\n",
       " 'upset',\n",
       " 'putin',\n",
       " 'forc',\n",
       " 'hurt',\n",
       " 'everybodi',\n",
       " 'tsybulskaliubov',\n",
       " 'cyka',\n",
       " 'blyat',\n",
       " 'give',\n",
       " 'sugar',\n",
       " 'die']"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "spam_corpus"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Counter({'ekaterina': 1,\n",
       "         'vinokourova': 1,\n",
       "         'daughter': 1,\n",
       "         'russian': 7,\n",
       "         'foreign': 1,\n",
       "         'minist': 1,\n",
       "         'sergei': 1,\n",
       "         'lavrov': 1,\n",
       "         'spent': 1,\n",
       "         'year': 2,\n",
       "         'u': 1,\n",
       "         'studi': 1,\n",
       "         'polit': 1,\n",
       "         'scienc': 1,\n",
       "         'columbia': 1,\n",
       "         'univers': 1,\n",
       "         'econom': 2,\n",
       "         'london': 1,\n",
       "         'school': 1,\n",
       "         'cyka': 31,\n",
       "         'go': 4,\n",
       "         'back': 1,\n",
       "         'motherfuckingruissia': 1,\n",
       "         'http': 5,\n",
       "         'co': 5,\n",
       "         'znkcrdqazz': 1,\n",
       "         'eye': 1,\n",
       "         'naught': 1,\n",
       "         'want': 1,\n",
       "         'two': 1,\n",
       "         'hear': 1,\n",
       "         'blyat': 15,\n",
       "         'ew': 1,\n",
       "         'fkgam': 1,\n",
       "         'gs': 1,\n",
       "         'xw': 1,\n",
       "         'rwapodcast': 1,\n",
       "         'bro': 2,\n",
       "         'supper': 1,\n",
       "         'offens': 1,\n",
       "         'tell': 1,\n",
       "         'also': 2,\n",
       "         'call': 1,\n",
       "         'bori': 1,\n",
       "         'channel': 1,\n",
       "         'ye': 1,\n",
       "         'dab': 1,\n",
       "         'hard': 1,\n",
       "         'yeah': 1,\n",
       "         'sure': 1,\n",
       "         'dub': 1,\n",
       "         'right': 1,\n",
       "         'true': 2,\n",
       "         'mememan': 1,\n",
       "         'cryptic': 4,\n",
       "         'hmmmmmmmmm': 1,\n",
       "         'women': 1,\n",
       "         'hmmm': 1,\n",
       "         'realli': 2,\n",
       "         'good': 1,\n",
       "         'draw': 1,\n",
       "         'man': 1,\n",
       "         'know': 2,\n",
       "         'deal': 1,\n",
       "         'teacher': 1,\n",
       "         'didnt': 1,\n",
       "         'wipe': 1,\n",
       "         'sinc': 1,\n",
       "         'friday': 1,\n",
       "         'fridaynightfunkin': 1,\n",
       "         'eorhjtxhsi': 1,\n",
       "         'aregularnam': 1,\n",
       "         'wait': 1,\n",
       "         'mean': 3,\n",
       "         'thing': 1,\n",
       "         'id': 1,\n",
       "         'let': 1,\n",
       "         'gnzyv': 1,\n",
       "         'tcq': 1,\n",
       "         'god': 1,\n",
       "         'lt': 1,\n",
       "         'yepyep': 1,\n",
       "         'nigellovesy': 1,\n",
       "         'soooo': 1,\n",
       "         'overr': 1,\n",
       "         'fuck': 6,\n",
       "         'noob': 1,\n",
       "         'game': 3,\n",
       "         'follow': 1,\n",
       "         'namastekikii': 1,\n",
       "         'liter': 1,\n",
       "         'bitch': 1,\n",
       "         'idiomat': 1,\n",
       "         'use': 2,\n",
       "         'stand': 1,\n",
       "         'sort': 1,\n",
       "         'explet': 1,\n",
       "         'english': 1,\n",
       "         'googl': 1,\n",
       "         'interest': 1,\n",
       "         'cykaart': 1,\n",
       "         'niceee': 1,\n",
       "         'congrat': 1,\n",
       "         'deserv': 1,\n",
       "         'tagaan': 1,\n",
       "         'kog': 1,\n",
       "         'chanc': 1,\n",
       "         'mag': 1,\n",
       "         'learn': 1,\n",
       "         'ug': 1,\n",
       "         'new': 1,\n",
       "         'languag': 1,\n",
       "         'pick': 1,\n",
       "         'german': 1,\n",
       "         'bahug': 1,\n",
       "         'bylat': 1,\n",
       "         'defenceu': 1,\n",
       "         'sad': 1,\n",
       "         'pathet': 1,\n",
       "         'putin': 3,\n",
       "         'need': 3,\n",
       "         'help': 2,\n",
       "         'china': 1,\n",
       "         'mercenari': 2,\n",
       "         'syria': 1,\n",
       "         'attack': 1,\n",
       "         'ukrain': 3,\n",
       "         'littl': 1,\n",
       "         'glori': 1,\n",
       "         'taliaa': 1,\n",
       "         'novaa': 1,\n",
       "         'shut': 1,\n",
       "         'nm': 1,\n",
       "         'mfa': 2,\n",
       "         'russia': 1,\n",
       "         'rusembukrain': 1,\n",
       "         'osc': 2,\n",
       "         'rf': 1,\n",
       "         'russiaun': 1,\n",
       "         'dpol': 1,\n",
       "         'un': 2,\n",
       "         'rusembusa': 1,\n",
       "         'look': 2,\n",
       "         'like': 2,\n",
       "         'gentli': 1,\n",
       "         'place': 1,\n",
       "         'damag': 1,\n",
       "         'ground': 1,\n",
       "         'surround': 1,\n",
       "         'area': 1,\n",
       "         'nobodi': 1,\n",
       "         'believ': 1,\n",
       "         'american': 1,\n",
       "         'lmao': 1,\n",
       "         'say': 3,\n",
       "         'shit': 2,\n",
       "         'ass': 2,\n",
       "         'bird': 1,\n",
       "         'app': 1,\n",
       "         'time': 1,\n",
       "         'moreso': 1,\n",
       "         'dont': 1,\n",
       "         'trust': 1,\n",
       "         'peopl': 2,\n",
       "         'whose': 1,\n",
       "         'first': 1,\n",
       "         'instinct': 1,\n",
       "         'see': 1,\n",
       "         'someth': 1,\n",
       "         'dislik': 1,\n",
       "         'rage': 1,\n",
       "         'work': 1,\n",
       "         'class': 1,\n",
       "         'leadership': 1,\n",
       "         'take': 1,\n",
       "         'sec': 1,\n",
       "         'paus': 1,\n",
       "         'think': 1,\n",
       "         'move': 1,\n",
       "         'advanc': 1,\n",
       "         'thank': 2,\n",
       "         'yuri': 1,\n",
       "         'teach': 1,\n",
       "         'phrase': 1,\n",
       "         'today': 1,\n",
       "         'learnt': 1,\n",
       "         'friendship': 1,\n",
       "         'tex': 1,\n",
       "         'mqmt': 1,\n",
       "         'marketrebel': 1,\n",
       "         'sooner': 1,\n",
       "         'later': 1,\n",
       "         'server': 1,\n",
       "         'gon': 1,\n",
       "         'na': 1,\n",
       "         'full': 1,\n",
       "         'kid': 1,\n",
       "         'scream': 1,\n",
       "         'blyattttt': 1,\n",
       "         'katiewonn': 1,\n",
       "         'hi': 1,\n",
       "         'friend': 1,\n",
       "         'name': 1,\n",
       "         'maxward': 1,\n",
       "         'pleas': 2,\n",
       "         'ignor': 1,\n",
       "         'messag': 1,\n",
       "         'enough': 1,\n",
       "         'whatsapp': 1,\n",
       "         'number': 1,\n",
       "         'rogozin': 1,\n",
       "         'profil': 1,\n",
       "         'pic': 1,\n",
       "         'budget': 1,\n",
       "         'photoshop': 1,\n",
       "         'caus': 1,\n",
       "         'head': 1,\n",
       "         'way': 1,\n",
       "         'big': 1,\n",
       "         'fat': 1,\n",
       "         'photo': 1,\n",
       "         'fake': 1,\n",
       "         'brokenboi': 1,\n",
       "         'get': 1,\n",
       "         'absolut': 1,\n",
       "         'own': 1,\n",
       "         'stupid': 1,\n",
       "         'support': 1,\n",
       "         'nazi': 1,\n",
       "         'cunt': 1,\n",
       "         'post': 1,\n",
       "         'block': 1,\n",
       "         'decid': 1,\n",
       "         'deactiv': 1,\n",
       "         'melon': 1,\n",
       "         'f': 1,\n",
       "         'c': 1,\n",
       "         'hywsyu': 1,\n",
       "         'barackobama': 1,\n",
       "         'eat': 1,\n",
       "         'ubercsgo': 1,\n",
       "         'timjdillon': 1,\n",
       "         'kiev': 1,\n",
       "         'fight': 1,\n",
       "         'gg': 1,\n",
       "         'toosefn': 1,\n",
       "         'elonmusk': 1,\n",
       "         'challeng': 1,\n",
       "         'cs': 1,\n",
       "         'match': 1,\n",
       "         'could': 1,\n",
       "         'shout': 1,\n",
       "         'teabag': 1,\n",
       "         'corps': 1,\n",
       "         'eigenrobot': 1,\n",
       "         'even': 2,\n",
       "         'heartbreak': 1,\n",
       "         'ig': 1,\n",
       "         'gen': 1,\n",
       "         'z': 1,\n",
       "         'front': 1,\n",
       "         'line': 2,\n",
       "         'surli': 1,\n",
       "         'prod': 1,\n",
       "         'forward': 1,\n",
       "         'els': 1,\n",
       "         'make': 1,\n",
       "         'upset': 1,\n",
       "         'forc': 1,\n",
       "         'hurt': 1,\n",
       "         'everybodi': 1,\n",
       "         'tsybulskaliubov': 1,\n",
       "         'give': 1,\n",
       "         'sugar': 1,\n",
       "         'die': 1})"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from collections import Counter\n",
    "Counter(spam_corpus)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>cyka</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>blyat</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>russian</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>fuck</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>http</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>co</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>go</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>cryptic</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>mean</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>game</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>putin</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>need</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>ukrain</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>say</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>year</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>econom</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>bro</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>also</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>true</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>realli</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>know</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>use</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>help</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>mercenari</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>mfa</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>osc</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>un</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>look</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>like</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>shit</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            0   1\n",
       "0        cyka  31\n",
       "1       blyat  15\n",
       "2     russian   7\n",
       "3        fuck   6\n",
       "4        http   5\n",
       "5          co   5\n",
       "6          go   4\n",
       "7     cryptic   4\n",
       "8        mean   3\n",
       "9        game   3\n",
       "10      putin   3\n",
       "11       need   3\n",
       "12     ukrain   3\n",
       "13        say   3\n",
       "14       year   2\n",
       "15     econom   2\n",
       "16        bro   2\n",
       "17       also   2\n",
       "18       true   2\n",
       "19     realli   2\n",
       "20       know   2\n",
       "21        use   2\n",
       "22       help   2\n",
       "23  mercenari   2\n",
       "24        mfa   2\n",
       "25        osc   2\n",
       "26         un   2\n",
       "27       look   2\n",
       "28       like   2\n",
       "29       shit   2"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.DataFrame(Counter(spam_corpus).most_common(30))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Chacha Tech\\anaconda3\\lib\\site-packages\\seaborn\\_decorators.py:36: FutureWarning: Pass the following variables as keyword args: x, y. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12,6))\n",
    "sns.barplot(pd.DataFrame(Counter(spam_corpus).most_common(30))[0], pd.DataFrame(Counter(spam_corpus).most_common(30))[1])\n",
    "plt.xlabel('Most Frequent Words')\n",
    "plt.ylabel('Count')\n",
    "plt.xticks(rotation='vertical')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "sentences = df['transformed_tweets'].tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "289"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(sentences)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "sentences_as_one_string = \" \".join(sentences)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x12f645cf1f0>"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (20, 20))\n",
    "plt.imshow(WordCloud().generate(sentences_as_one_string))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# visualize the value counts.\n",
    "plt.figure(figsize=(10,4))\n",
    "df['analysis'].value_counts().plot(kind='bar')\n",
    "plt.xlabel('sentiment')\n",
    "plt.ylabel('count')\n",
    "plt.title('Sentiment Analysis')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['new_target'] = df['transformed_tweets'].apply(lambda x: 1 if 'cyka' in x or 'blyat' in x or 'fuck' in x else 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "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>tweets</th>\n",
       "      <th>target</th>\n",
       "      <th>num_of_characters</th>\n",
       "      <th>num_of_words</th>\n",
       "      <th>num_of_sentences</th>\n",
       "      <th>transformed_tweets</th>\n",
       "      <th>subjectivity</th>\n",
       "      <th>polarity</th>\n",
       "      <th>analysis</th>\n",
       "      <th>new_target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ekaterina Vinokourova  the only daughter of Ru...</td>\n",
       "      <td>1</td>\n",
       "      <td>303</td>\n",
       "      <td>46</td>\n",
       "      <td>1</td>\n",
       "      <td>ekaterina vinokourova daughter russian foreign...</td>\n",
       "      <td>0.056250</td>\n",
       "      <td>-0.05625</td>\n",
       "      <td>negative</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>eye do naught want two hear it  cyka blyat</td>\n",
       "      <td>1</td>\n",
       "      <td>42</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>eye naught want two hear cyka blyat</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...</td>\n",
       "      <td>1</td>\n",
       "      <td>221</td>\n",
       "      <td>41</td>\n",
       "      <td>1</td>\n",
       "      <td>ew fkgam gs xw rwapodcast cyka bro supper offe...</td>\n",
       "      <td>0.523254</td>\n",
       "      <td>0.16881</td>\n",
       "      <td>positive</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cryptic Cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>cryptic cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Cryptic Cyka that s a really good drawing man</td>\n",
       "      <td>1</td>\n",
       "      <td>46</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>cryptic cyka realli good draw man</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.70000</td>\n",
       "      <td>positive</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              tweets  target  \\\n",
       "0  Ekaterina Vinokourova  the only daughter of Ru...       1   \n",
       "1         eye do naught want two hear it  cyka blyat       1   \n",
       "2   eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...       1   \n",
       "3                 Cryptic Cyka hmmmmmmmmm women hmmm       1   \n",
       "4      Cryptic Cyka that s a really good drawing man       1   \n",
       "\n",
       "   num_of_characters  num_of_words  num_of_sentences  \\\n",
       "0                303            46                 1   \n",
       "1                 42             9                 1   \n",
       "2                221            41                 1   \n",
       "3                 35             5                 1   \n",
       "4                 46             9                 1   \n",
       "\n",
       "                                  transformed_tweets  subjectivity  polarity  \\\n",
       "0  ekaterina vinokourova daughter russian foreign...      0.056250  -0.05625   \n",
       "1                eye naught want two hear cyka blyat      0.000000   0.00000   \n",
       "2  ew fkgam gs xw rwapodcast cyka bro supper offe...      0.523254   0.16881   \n",
       "3                 cryptic cyka hmmmmmmmmm women hmmm      0.000000   0.00000   \n",
       "4                  cryptic cyka realli good draw man      0.600000   0.70000   \n",
       "\n",
       "   analysis  new_target  \n",
       "0  negative           1  \n",
       "1   neutral           1  \n",
       "2  positive           1  \n",
       "3   neutral           1  \n",
       "4  positive           1  "
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "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>tweets</th>\n",
       "      <th>target</th>\n",
       "      <th>num_of_characters</th>\n",
       "      <th>num_of_words</th>\n",
       "      <th>num_of_sentences</th>\n",
       "      <th>transformed_tweets</th>\n",
       "      <th>subjectivity</th>\n",
       "      <th>polarity</th>\n",
       "      <th>analysis</th>\n",
       "      <th>new_target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>290</th>\n",
       "      <td>I love thi game</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>love thi game</td>\n",
       "      <td>0.50</td>\n",
       "      <td>0.05</td>\n",
       "      <td>positive</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>291</th>\n",
       "      <td>game graphics is eye catching</td>\n",
       "      <td>0</td>\n",
       "      <td>30</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>game graphic eye catch</td>\n",
       "      <td>0.40</td>\n",
       "      <td>-0.20</td>\n",
       "      <td>negative</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>292</th>\n",
       "      <td>Game level is really enjoyable</td>\n",
       "      <td>0</td>\n",
       "      <td>31</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>game level realli enjoy</td>\n",
       "      <td>0.45</td>\n",
       "      <td>0.00</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>293</th>\n",
       "      <td>Good Pixel and very beautiful tool</td>\n",
       "      <td>0</td>\n",
       "      <td>35</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>good pixel beauti tool</td>\n",
       "      <td>0.60</td>\n",
       "      <td>0.70</td>\n",
       "      <td>positive</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>294</th>\n",
       "      <td>you are free so come online and play with me</td>\n",
       "      <td>0</td>\n",
       "      <td>45</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>free come onlin play</td>\n",
       "      <td>0.80</td>\n",
       "      <td>0.40</td>\n",
       "      <td>positive</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            tweets  target  num_of_characters  \\\n",
       "290                                I love thi game       0                 16   \n",
       "291                  game graphics is eye catching       0                 30   \n",
       "292                 Game level is really enjoyable       0                 31   \n",
       "293             Good Pixel and very beautiful tool       0                 35   \n",
       "294   you are free so come online and play with me       0                 45   \n",
       "\n",
       "     num_of_words  num_of_sentences       transformed_tweets  subjectivity  \\\n",
       "290             4                 1            love thi game          0.50   \n",
       "291             5                 1   game graphic eye catch          0.40   \n",
       "292             5                 1  game level realli enjoy          0.45   \n",
       "293             6                 1   good pixel beauti tool          0.60   \n",
       "294            10                 1     free come onlin play          0.80   \n",
       "\n",
       "     polarity  analysis  new_target  \n",
       "290      0.05  positive           0  \n",
       "291     -0.20  negative           0  \n",
       "292      0.00   neutral           0  \n",
       "293      0.70  positive           0  \n",
       "294      0.40  positive           0  "
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    243\n",
       "1     46\n",
       "Name: new_target, dtype: int64"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['new_target'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['cyka'] = df['transformed_tweets'].apply(lambda x: 1 if 'cyka' in x else 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['blyat'] = df['transformed_tweets'].apply(lambda x: 1 if 'blyat' in x else 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['fuck'] = df['transformed_tweets'].apply(lambda x: 1 if 'fuck' in x else 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    245\n",
       "1     44\n",
       "Name: cyka, dtype: int64"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['cyka'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    264\n",
       "1     25\n",
       "Name: blyat, dtype: int64"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['blyat'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    280\n",
       "1      9\n",
       "Name: fuck, dtype: int64"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['fuck'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Model Building"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "# naive based machine learning is best model for textual data.\n",
    " \n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "cv = CountVectorizer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = cv.fit_transform(df['transformed_tweets']).toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0, 0, 0, ..., 1, 0, 0],\n",
       "       [0, 0, 0, ..., 0, 0, 0],\n",
       "       [0, 0, 0, ..., 0, 0, 0],\n",
       "       ...,\n",
       "       [0, 0, 0, ..., 0, 0, 0],\n",
       "       [0, 0, 0, ..., 0, 0, 0],\n",
       "       [0, 0, 0, ..., 0, 0, 0]], dtype=int64)"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(289, 1531)"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = df['new_target'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(289,)"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.naive_bayes import GaussianNB, MultinomialNB, BernoulliNB\n",
    "from sklearn.metrics import accuracy_score, confusion_matrix, precision_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "gnb = GaussianNB()\n",
    "mnb = MultinomialNB()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "# # imbalanced data me accuracy matter zyada nh krta lekin precision score zyada se zyada ho."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "gnb.fit(X_train, y_train)\n",
    "y_pred1 = gnb.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy Score :  0.9310344827586207\n",
      "Confusion Matrix:  [[46  3]\n",
      " [ 1  8]]\n",
      "Precision Score :  0.7272727272727273\n"
     ]
    }
   ],
   "source": [
    "print('Accuracy Score : ', accuracy_score(y_test, y_pred1))\n",
    "print('Confusion Matrix: ', confusion_matrix(y_test, y_pred1))\n",
    "print('Precision Score : ', precision_score(y_test, y_pred1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "mnb.fit(X_train, y_train)\n",
    "y_pred2 = mnb.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy Score :  0.896551724137931\n",
      "Confusion Matrix:  [[43  6]\n",
      " [ 0  9]]\n",
      "Precision Score :  0.6\n"
     ]
    }
   ],
   "source": [
    "print('Accuracy Score : ', accuracy_score(y_test, y_pred2))\n",
    "print('Confusion Matrix: ', confusion_matrix(y_test, y_pred2))\n",
    "print('Precision Score : ', precision_score(y_test, y_pred2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "# TFidf vectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "tfidf = TfidfVectorizer(max_features = 3000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_tfidf = tfidf.fit_transform(df['transformed_tweets']).toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.        , 0.        , 0.        , ..., 0.19876558, 0.        ,\n",
       "        0.        ],\n",
       "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
       "        0.        ],\n",
       "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
       "        0.        ],\n",
       "       ...,\n",
       "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
       "        0.        ],\n",
       "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
       "        0.        ],\n",
       "       [0.        , 0.        , 0.        , ..., 0.        , 0.        ,\n",
       "        0.        ]])"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_tfidf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(289, 1531)"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_tfidf.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(289,)"
      ]
     },
     "execution_count": 96,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X_tfidf, y, test_size = 0.2, random_state = 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "gnb_tfidf = GaussianNB()\n",
    "mnb_tfidf = MultinomialNB()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [],
   "source": [
    "gnb_tfidf.fit(X_train, y_train)\n",
    "y_pred4 = gnb_tfidf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy Score :  0.896551724137931\n",
      "Confusion Matrix:  [[46  3]\n",
      " [ 3  6]]\n",
      "Precision Score :  0.6666666666666666\n"
     ]
    }
   ],
   "source": [
    "print('Accuracy Score : ', accuracy_score(y_test, y_pred4))\n",
    "print('Confusion Matrix: ', confusion_matrix(y_test, y_pred4))\n",
    "print('Precision Score : ', precision_score(y_test, y_pred4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [],
   "source": [
    "mnb_tfidf.fit(X_train, y_train)\n",
    "y_pred5 = mnb_tfidf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy Score :  0.8793103448275862\n",
      "Confusion Matrix:  [[49  0]\n",
      " [ 7  2]]\n",
      "Precision Score :  1.0\n"
     ]
    }
   ],
   "source": [
    "print('Accuracy Score : ', accuracy_score(y_test, y_pred5))\n",
    "print('Confusion Matrix: ', confusion_matrix(y_test, y_pred5))\n",
    "print('Precision Score : ', precision_score(y_test, y_pred5))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "bnb_tfidf = BernoulliNB()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [],
   "source": [
    "bnb_tfidf.fit(X_train, y_train)\n",
    "y_pred6 = bnb_tfidf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy Score :  0.8275862068965517\n",
      "Confusion Matrix:  [[48  1]\n",
      " [ 9  0]]\n",
      "Precision Score :  0.0\n"
     ]
    }
   ],
   "source": [
    "print('Accuracy Score : ', accuracy_score(y_test, y_pred6))\n",
    "print('Confusion Matrix: ', confusion_matrix(y_test, y_pred6))\n",
    "print('Precision Score : ', precision_score(y_test, y_pred6))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "# let's compare other models one by one"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.svm import SVC\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.ensemble import AdaBoostClassifier\n",
    "from sklearn.ensemble import BaggingClassifier\n",
    "from sklearn.ensemble import ExtraTreesClassifier\n",
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "from xgboost import XGBClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [],
   "source": [
    "svc = SVC(kernel = 'sigmoid', gamma = 1.0)\n",
    "knc = KNeighborsClassifier()\n",
    "mnb = MultinomialNB()\n",
    "dtc = DecisionTreeClassifier(max_depth = 5)\n",
    "lrc = LogisticRegression(solver = 'liblinear', penalty = 'l1')\n",
    "rfc = RandomForestClassifier(n_estimators = 50, random_state = 2)\n",
    "abc = AdaBoostClassifier(n_estimators = 50, random_state = 2)\n",
    "bc  = BaggingClassifier(n_estimators = 50, random_state = 2)\n",
    "gbc = GradientBoostingClassifier(n_estimators = 50, random_state = 2)\n",
    "xgb = XGBClassifier(n_estimators = 50, random_state = 2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "clfs = {\n",
    "    'SVC' : svc,\n",
    "    'KNC' : knc,\n",
    "    'MNB' : mnb,\n",
    "    'DTC' : dtc,\n",
    "    'LRC' : lrc,\n",
    "    'RFC' : rfc,\n",
    "    'ABC' : abc,\n",
    "    'BC'  : bc,\n",
    "    'GBC' : gbc,\n",
    "    'XGB' : xgb\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [],
   "source": [
    "def train_test_classifier(clf, X_train, y_train,  X_test, y_test):\n",
    "    clf.fit(X_train, y_train)\n",
    "    y_pred = clf.predict(X_test)\n",
    "    accuracy  = accuracy_score(y_test, y_pred)\n",
    "    precision = precision_score(y_test, y_pred)\n",
    "    \n",
    "    return accuracy, precision"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.9827586206896551, 1.0)"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_test_classifier(svc, X_train, y_train, X_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "For  SVC\n",
      "Accuracy -  0.9827586206896551\n",
      "Precision -  1.0\n",
      "For  KNC\n",
      "Accuracy -  1.0\n",
      "Precision -  1.0\n",
      "For  MNB\n",
      "Accuracy -  0.8793103448275862\n",
      "Precision -  1.0\n",
      "For  DTC\n",
      "Accuracy -  1.0\n",
      "Precision -  1.0\n",
      "For  LRC\n",
      "Accuracy -  1.0\n",
      "Precision -  1.0\n",
      "For  RFC\n",
      "Accuracy -  0.9827586206896551\n",
      "Precision -  1.0\n",
      "For  ABC\n",
      "Accuracy -  1.0\n",
      "Precision -  1.0\n",
      "For  BC\n",
      "Accuracy -  1.0\n",
      "Precision -  1.0\n",
      "For  GBC\n",
      "Accuracy -  1.0\n",
      "Precision -  1.0\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Chacha Tech\\anaconda3\\lib\\site-packages\\xgboost\\sklearn.py:1146: UserWarning: The use of label encoder in XGBClassifier is deprecated and will be removed in a future release. To remove this warning, do the following: 1) Pass option use_label_encoder=False when constructing XGBClassifier object; and 2) Encode your labels (y) as integers starting with 0, i.e. 0, 1, 2, ..., [num_class - 1].\n",
      "  warnings.warn(label_encoder_deprecation_msg, UserWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[21:05:01] WARNING: C:/Users/Administrator/workspace/xgboost-win64_release_1.4.0/src/learner.cc:1095: Starting in XGBoost 1.3.0, the default evaluation metric used with the objective 'binary:logistic' was changed from 'error' to 'logloss'. Explicitly set eval_metric if you'd like to restore the old behavior.\n",
      "For  XGB\n",
      "Accuracy -  1.0\n",
      "Precision -  1.0\n"
     ]
    }
   ],
   "source": [
    "accuracy_scores  = []\n",
    "precision_scores = []\n",
    "\n",
    "for name, clf in clfs.items():\n",
    "    current_accuracy, current_precision = train_test_classifier(clf, X_train, y_train,  X_test, y_test)\n",
    "    \n",
    "    print(\"For \", name)\n",
    "    print(\"Accuracy - \", current_accuracy)\n",
    "    print(\"Precision - \", current_precision)\n",
    "    \n",
    "    accuracy_scores.append(current_accuracy)\n",
    "    precision_scores.append(current_precision)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "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>Algorithm</th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>Precision</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SVC</td>\n",
       "      <td>0.982759</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>KNC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>MNB</td>\n",
       "      <td>0.879310</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DTC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LRC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>RFC</td>\n",
       "      <td>0.982759</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>ABC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>BC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>GBC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>XGB</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Algorithm  Accuracy  Precision\n",
       "0       SVC  0.982759        1.0\n",
       "1       KNC  1.000000        1.0\n",
       "2       MNB  0.879310        1.0\n",
       "3       DTC  1.000000        1.0\n",
       "4       LRC  1.000000        1.0\n",
       "5       RFC  0.982759        1.0\n",
       "6       ABC  1.000000        1.0\n",
       "7        BC  1.000000        1.0\n",
       "8       GBC  1.000000        1.0\n",
       "9       XGB  1.000000        1.0"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "performance_df = pd.DataFrame({'Algorithm':clfs.keys(),'Accuracy':accuracy_scores,'Precision':precision_scores})\n",
    "performance_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "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>Algorithm</th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>Precision</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>KNC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DTC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LRC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>ABC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>BC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>GBC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>XGB</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SVC</td>\n",
       "      <td>0.982759</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>RFC</td>\n",
       "      <td>0.982759</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>MNB</td>\n",
       "      <td>0.879310</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Algorithm  Accuracy  Precision\n",
       "1       KNC  1.000000        1.0\n",
       "3       DTC  1.000000        1.0\n",
       "4       LRC  1.000000        1.0\n",
       "6       ABC  1.000000        1.0\n",
       "7        BC  1.000000        1.0\n",
       "8       GBC  1.000000        1.0\n",
       "9       XGB  1.000000        1.0\n",
       "0       SVC  0.982759        1.0\n",
       "5       RFC  0.982759        1.0\n",
       "2       MNB  0.879310        1.0"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "performance_df_sv_acc = pd.DataFrame({'Algorithm':clfs.keys(),'Accuracy':accuracy_scores,'Precision':precision_scores}).sort_values('Accuracy',ascending=False)\n",
    "performance_df_sv_acc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [],
   "source": [
    "performance_df = pd.DataFrame({'Algorithm':clfs.keys(),'Accuracy':accuracy_scores,'Precision':precision_scores}).sort_values('Precision',ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "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>Algorithm</th>\n",
       "      <th>Accuracy</th>\n",
       "      <th>Precision</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SVC</td>\n",
       "      <td>0.982759</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>KNC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>MNB</td>\n",
       "      <td>0.879310</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DTC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LRC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>RFC</td>\n",
       "      <td>0.982759</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>ABC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>BC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>GBC</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>XGB</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Algorithm  Accuracy  Precision\n",
       "0       SVC  0.982759        1.0\n",
       "1       KNC  1.000000        1.0\n",
       "2       MNB  0.879310        1.0\n",
       "3       DTC  1.000000        1.0\n",
       "4       LRC  1.000000        1.0\n",
       "5       RFC  0.982759        1.0\n",
       "6       ABC  1.000000        1.0\n",
       "7        BC  1.000000        1.0\n",
       "8       GBC  1.000000        1.0\n",
       "9       XGB  1.000000        1.0"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "performance_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "performance_df1 = pd.melt(performance_df, id_vars = 'Algorithm')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "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>Algorithm</th>\n",
       "      <th>variable</th>\n",
       "      <th>value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SVC</td>\n",
       "      <td>Accuracy</td>\n",
       "      <td>0.982759</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>KNC</td>\n",
       "      <td>Accuracy</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>MNB</td>\n",
       "      <td>Accuracy</td>\n",
       "      <td>0.879310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DTC</td>\n",
       "      <td>Accuracy</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>LRC</td>\n",
       "      <td>Accuracy</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>RFC</td>\n",
       "      <td>Accuracy</td>\n",
       "      <td>0.982759</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>ABC</td>\n",
       "      <td>Accuracy</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>BC</td>\n",
       "      <td>Accuracy</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>GBC</td>\n",
       "      <td>Accuracy</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>XGB</td>\n",
       "      <td>Accuracy</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>SVC</td>\n",
       "      <td>Precision</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>KNC</td>\n",
       "      <td>Precision</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>MNB</td>\n",
       "      <td>Precision</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>DTC</td>\n",
       "      <td>Precision</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>LRC</td>\n",
       "      <td>Precision</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>RFC</td>\n",
       "      <td>Precision</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>ABC</td>\n",
       "      <td>Precision</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>BC</td>\n",
       "      <td>Precision</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>GBC</td>\n",
       "      <td>Precision</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>XGB</td>\n",
       "      <td>Precision</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Algorithm   variable     value\n",
       "0        SVC   Accuracy  0.982759\n",
       "1        KNC   Accuracy  1.000000\n",
       "2        MNB   Accuracy  0.879310\n",
       "3        DTC   Accuracy  1.000000\n",
       "4        LRC   Accuracy  1.000000\n",
       "5        RFC   Accuracy  0.982759\n",
       "6        ABC   Accuracy  1.000000\n",
       "7         BC   Accuracy  1.000000\n",
       "8        GBC   Accuracy  1.000000\n",
       "9        XGB   Accuracy  1.000000\n",
       "10       SVC  Precision  1.000000\n",
       "11       KNC  Precision  1.000000\n",
       "12       MNB  Precision  1.000000\n",
       "13       DTC  Precision  1.000000\n",
       "14       LRC  Precision  1.000000\n",
       "15       RFC  Precision  1.000000\n",
       "16       ABC  Precision  1.000000\n",
       "17        BC  Precision  1.000000\n",
       "18       GBC  Precision  1.000000\n",
       "19       XGB  Precision  1.000000"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "performance_df1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1440x576 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 441.75x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (20, 8))\n",
    "sns.catplot(x = 'Algorithm', y = 'value', hue = 'variable', data = performance_df1, kind = 'bar', height = 5)\n",
    "\n",
    "plt.ylim(0.5, 1.0)\n",
    "plt.xticks(rotation = 'vertical')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Voting Classifier\n",
    "svc = SVC(kernel='sigmoid', gamma=1.0,probability=True)\n",
    "mnb = MultinomialNB()\n",
    "etc = ExtraTreesClassifier(n_estimators=50, random_state=2)\n",
    "\n",
    "from sklearn.ensemble import VotingClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [],
   "source": [
    "voting = VotingClassifier(estimators=[('svm', svc), ('nb', mnb), ('et', etc)],voting='soft')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "VotingClassifier(estimators=[('svm',\n",
       "                              SVC(gamma=1.0, kernel='sigmoid',\n",
       "                                  probability=True)),\n",
       "                             ('nb', MultinomialNB()),\n",
       "                             ('et',\n",
       "                              ExtraTreesClassifier(n_estimators=50,\n",
       "                                                   random_state=2))],\n",
       "                 voting='soft')"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "voting.fit(X_train,y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy 1.0\n",
      "Precision 1.0\n"
     ]
    }
   ],
   "source": [
    "y_pred = voting.predict(X_test)\n",
    "print(\"Accuracy\",accuracy_score(y_test,y_pred))\n",
    "print(\"Precision\",precision_score(y_test,y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Applying stacking\n",
    "estimators=[('svm', svc), ('nb', mnb), ('et', etc)]\n",
    "final_estimator=RandomForestClassifier()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import StackingClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = StackingClassifier(estimators=estimators, final_estimator=final_estimator)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy 0.9827586206896551\n",
      "Precision 0.9\n"
     ]
    }
   ],
   "source": [
    "clf.fit(X_train,y_train)\n",
    "y_pred = clf.predict(X_test)\n",
    "print(\"Accuracy\",accuracy_score(y_test,y_pred))\n",
    "print(\"Precision\",precision_score(y_test,y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "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>tweets</th>\n",
       "      <th>target</th>\n",
       "      <th>num_of_characters</th>\n",
       "      <th>num_of_words</th>\n",
       "      <th>num_of_sentences</th>\n",
       "      <th>transformed_tweets</th>\n",
       "      <th>subjectivity</th>\n",
       "      <th>polarity</th>\n",
       "      <th>analysis</th>\n",
       "      <th>new_target</th>\n",
       "      <th>cyka</th>\n",
       "      <th>blyat</th>\n",
       "      <th>fuck</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ekaterina Vinokourova  the only daughter of Ru...</td>\n",
       "      <td>1</td>\n",
       "      <td>303</td>\n",
       "      <td>46</td>\n",
       "      <td>1</td>\n",
       "      <td>ekaterina vinokourova daughter russian foreign...</td>\n",
       "      <td>0.056250</td>\n",
       "      <td>-0.05625</td>\n",
       "      <td>negative</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>eye do naught want two hear it  cyka blyat</td>\n",
       "      <td>1</td>\n",
       "      <td>42</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>eye naught want two hear cyka blyat</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...</td>\n",
       "      <td>1</td>\n",
       "      <td>221</td>\n",
       "      <td>41</td>\n",
       "      <td>1</td>\n",
       "      <td>ew fkgam gs xw rwapodcast cyka bro supper offe...</td>\n",
       "      <td>0.523254</td>\n",
       "      <td>0.16881</td>\n",
       "      <td>positive</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cryptic Cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>1</td>\n",
       "      <td>35</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>cryptic cyka hmmmmmmmmm women hmmm</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Cryptic Cyka that s a really good drawing man</td>\n",
       "      <td>1</td>\n",
       "      <td>46</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>cryptic cyka realli good draw man</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>0.70000</td>\n",
       "      <td>positive</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Cryptic Cyka they knows her deal</td>\n",
       "      <td>1</td>\n",
       "      <td>33</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>cryptic cyka know deal</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Teacher didnt wipe her off since friday  frida...</td>\n",
       "      <td>1</td>\n",
       "      <td>82</td>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "      <td>teacher didnt wipe sinc friday fridaynightfunk...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Aregularname Wait what does that mean</td>\n",
       "      <td>1</td>\n",
       "      <td>39</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>aregularnam wait mean</td>\n",
       "      <td>0.687500</td>\n",
       "      <td>-0.31250</td>\n",
       "      <td>negative</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>The things id let him do   https   t co gnZYv ...</td>\n",
       "      <td>1</td>\n",
       "      <td>50</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>thing id let http co gnzyv tcq</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Cryptic Cyka on god  lt</td>\n",
       "      <td>1</td>\n",
       "      <td>26</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>cryptic cyka god lt</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>neutral</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              tweets  target  \\\n",
       "0  Ekaterina Vinokourova  the only daughter of Ru...       1   \n",
       "1         eye do naught want two hear it  cyka blyat       1   \n",
       "2   eW fkgAM  GS xW  RWApodcast Cyka  Bro that s ...       1   \n",
       "3                 Cryptic Cyka hmmmmmmmmm women hmmm       1   \n",
       "4      Cryptic Cyka that s a really good drawing man       1   \n",
       "5                   Cryptic Cyka they knows her deal       1   \n",
       "6  Teacher didnt wipe her off since friday  frida...       1   \n",
       "7             Aregularname Wait what does that mean        1   \n",
       "8  The things id let him do   https   t co gnZYv ...       1   \n",
       "9                          Cryptic Cyka on god  lt         1   \n",
       "\n",
       "   num_of_characters  num_of_words  num_of_sentences  \\\n",
       "0                303            46                 1   \n",
       "1                 42             9                 1   \n",
       "2                221            41                 1   \n",
       "3                 35             5                 1   \n",
       "4                 46             9                 1   \n",
       "5                 33             6                 1   \n",
       "6                 82            12                 1   \n",
       "7                 39             6                 1   \n",
       "8                 50            11                 1   \n",
       "9                 26             5                 1   \n",
       "\n",
       "                                  transformed_tweets  subjectivity  polarity  \\\n",
       "0  ekaterina vinokourova daughter russian foreign...      0.056250  -0.05625   \n",
       "1                eye naught want two hear cyka blyat      0.000000   0.00000   \n",
       "2  ew fkgam gs xw rwapodcast cyka bro supper offe...      0.523254   0.16881   \n",
       "3                 cryptic cyka hmmmmmmmmm women hmmm      0.000000   0.00000   \n",
       "4                  cryptic cyka realli good draw man      0.600000   0.70000   \n",
       "5                             cryptic cyka know deal      0.000000   0.00000   \n",
       "6  teacher didnt wipe sinc friday fridaynightfunk...      0.000000   0.00000   \n",
       "7                              aregularnam wait mean      0.687500  -0.31250   \n",
       "8                     thing id let http co gnzyv tcq      0.000000   0.00000   \n",
       "9                                cryptic cyka god lt      0.000000   0.00000   \n",
       "\n",
       "   analysis  new_target  cyka  blyat  fuck  \n",
       "0  negative           1     1      0     1  \n",
       "1   neutral           1     1      1     0  \n",
       "2  positive           1     1      0     0  \n",
       "3   neutral           1     1      0     0  \n",
       "4  positive           1     1      0     0  \n",
       "5   neutral           1     1      0     0  \n",
       "6   neutral           0     0      0     0  \n",
       "7  negative           0     0      0     0  \n",
       "8   neutral           0     0      0     0  \n",
       "9   neutral           1     1      0     0  "
      ]
     },
     "execution_count": 128,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {},
   "outputs": [],
   "source": [
    "tweet = transform_text('eye cyka fuck do naught want two hear it ')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {},
   "outputs": [],
   "source": [
    "vector_input = tfidf.transform([tweet])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {},
   "outputs": [],
   "source": [
    "result = mnb_tfidf.predict(vector_input)[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {},
   "outputs": [],
   "source": [
    "def showresult(result):\n",
    "    if result == 1:\n",
    "        print('Toxic Tweet')\n",
    "    else:\n",
    "        print('Not Toxic Tweet')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Toxic Tweet\n"
     ]
    }
   ],
   "source": [
    "showresult(result)"
   ]
  },
  {
   "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}