sentiment_analysis_for_drug_reviews_with_models.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "090d6f72-7bbc-4778-911c-cad9f75f88a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import required libraries and modules\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "import warnings \n",
    "warnings.simplefilter('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ac491823-79b1-4d21-8a82-54f6294b05ca",
   "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>uniqueID</th>\n",
       "      <th>drugName</th>\n",
       "      <th>condition</th>\n",
       "      <th>review</th>\n",
       "      <th>rating</th>\n",
       "      <th>date</th>\n",
       "      <th>usefulCount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>206461</td>\n",
       "      <td>Valsartan</td>\n",
       "      <td>Left Ventricular Dysfunction</td>\n",
       "      <td>\"It has no side effect, I take it in combinati...</td>\n",
       "      <td>9</td>\n",
       "      <td>20-May-12</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>95260</td>\n",
       "      <td>Guanfacine</td>\n",
       "      <td>ADHD</td>\n",
       "      <td>\"My son is halfway through his fourth week of ...</td>\n",
       "      <td>8</td>\n",
       "      <td>27-Apr-10</td>\n",
       "      <td>192</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>92703</td>\n",
       "      <td>Lybrel</td>\n",
       "      <td>Birth Control</td>\n",
       "      <td>\"I used to take another oral contraceptive, wh...</td>\n",
       "      <td>5</td>\n",
       "      <td>14-Dec-09</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>138000</td>\n",
       "      <td>Ortho Evra</td>\n",
       "      <td>Birth Control</td>\n",
       "      <td>\"This is my first time using any form of birth...</td>\n",
       "      <td>8</td>\n",
       "      <td>3-Nov-15</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>35696</td>\n",
       "      <td>Buprenorphine / naloxone</td>\n",
       "      <td>Opiate Dependence</td>\n",
       "      <td>\"Suboxone has completely turned my life around...</td>\n",
       "      <td>9</td>\n",
       "      <td>27-Nov-16</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uniqueID                  drugName                     condition  \\\n",
       "0    206461                 Valsartan  Left Ventricular Dysfunction   \n",
       "1     95260                Guanfacine                          ADHD   \n",
       "2     92703                    Lybrel                 Birth Control   \n",
       "3    138000                Ortho Evra                 Birth Control   \n",
       "4     35696  Buprenorphine / naloxone             Opiate Dependence   \n",
       "\n",
       "                                              review  rating       date  \\\n",
       "0  \"It has no side effect, I take it in combinati...       9  20-May-12   \n",
       "1  \"My son is halfway through his fourth week of ...       8  27-Apr-10   \n",
       "2  \"I used to take another oral contraceptive, wh...       5  14-Dec-09   \n",
       "3  \"This is my first time using any form of birth...       8   3-Nov-15   \n",
       "4  \"Suboxone has completely turned my life around...       9  27-Nov-16   \n",
       "\n",
       "   usefulCount  \n",
       "0           27  \n",
       "1          192  \n",
       "2           17  \n",
       "3           10  \n",
       "4           37  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load the Train dataset\n",
    "train = pd.read_csv('train.csv')\n",
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "1ee40787-dc48-4f90-8f86-c66208950abe",
   "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>uniqueID</th>\n",
       "      <th>drugName</th>\n",
       "      <th>condition</th>\n",
       "      <th>review</th>\n",
       "      <th>rating</th>\n",
       "      <th>date</th>\n",
       "      <th>usefulCount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>163740</td>\n",
       "      <td>Mirtazapine</td>\n",
       "      <td>Depression</td>\n",
       "      <td>\"I&amp;#039;ve tried a few antidepressants over th...</td>\n",
       "      <td>10</td>\n",
       "      <td>28-Feb-12</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>206473</td>\n",
       "      <td>Mesalamine</td>\n",
       "      <td>Crohn's Disease, Maintenance</td>\n",
       "      <td>\"My son has Crohn&amp;#039;s disease and has done ...</td>\n",
       "      <td>8</td>\n",
       "      <td>17-May-09</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>159672</td>\n",
       "      <td>Bactrim</td>\n",
       "      <td>Urinary Tract Infection</td>\n",
       "      <td>\"Quick reduction of symptoms\"</td>\n",
       "      <td>9</td>\n",
       "      <td>29-Sep-17</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>39293</td>\n",
       "      <td>Contrave</td>\n",
       "      <td>Weight Loss</td>\n",
       "      <td>\"Contrave combines drugs that were used for al...</td>\n",
       "      <td>9</td>\n",
       "      <td>5-Mar-17</td>\n",
       "      <td>35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>97768</td>\n",
       "      <td>Cyclafem 1 / 35</td>\n",
       "      <td>Birth Control</td>\n",
       "      <td>\"I have been on this birth control for one cyc...</td>\n",
       "      <td>9</td>\n",
       "      <td>22-Oct-15</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uniqueID         drugName                     condition  \\\n",
       "0    163740      Mirtazapine                    Depression   \n",
       "1    206473       Mesalamine  Crohn's Disease, Maintenance   \n",
       "2    159672          Bactrim       Urinary Tract Infection   \n",
       "3     39293         Contrave                   Weight Loss   \n",
       "4     97768  Cyclafem 1 / 35                 Birth Control   \n",
       "\n",
       "                                              review  rating       date  \\\n",
       "0  \"I&#039;ve tried a few antidepressants over th...      10  28-Feb-12   \n",
       "1  \"My son has Crohn&#039;s disease and has done ...       8  17-May-09   \n",
       "2                      \"Quick reduction of symptoms\"       9  29-Sep-17   \n",
       "3  \"Contrave combines drugs that were used for al...       9   5-Mar-17   \n",
       "4  \"I have been on this birth control for one cyc...       9  22-Oct-15   \n",
       "\n",
       "   usefulCount  \n",
       "0           22  \n",
       "1           17  \n",
       "2            3  \n",
       "3           35  \n",
       "4            4  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load the Test dataset\n",
    "test = pd.read_csv('test.csv')\n",
    "test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "77749e65-d2a0-4c0e-b28a-79e456a4fb75",
   "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>uniqueID</th>\n",
       "      <th>drugName</th>\n",
       "      <th>condition</th>\n",
       "      <th>review</th>\n",
       "      <th>rating</th>\n",
       "      <th>date</th>\n",
       "      <th>usefulCount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>206461</td>\n",
       "      <td>Valsartan</td>\n",
       "      <td>Left Ventricular Dysfunction</td>\n",
       "      <td>\"It has no side effect, I take it in combinati...</td>\n",
       "      <td>9</td>\n",
       "      <td>20-May-12</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>95260</td>\n",
       "      <td>Guanfacine</td>\n",
       "      <td>ADHD</td>\n",
       "      <td>\"My son is halfway through his fourth week of ...</td>\n",
       "      <td>8</td>\n",
       "      <td>27-Apr-10</td>\n",
       "      <td>192</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>92703</td>\n",
       "      <td>Lybrel</td>\n",
       "      <td>Birth Control</td>\n",
       "      <td>\"I used to take another oral contraceptive, wh...</td>\n",
       "      <td>5</td>\n",
       "      <td>14-Dec-09</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>138000</td>\n",
       "      <td>Ortho Evra</td>\n",
       "      <td>Birth Control</td>\n",
       "      <td>\"This is my first time using any form of birth...</td>\n",
       "      <td>8</td>\n",
       "      <td>3-Nov-15</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>35696</td>\n",
       "      <td>Buprenorphine / naloxone</td>\n",
       "      <td>Opiate Dependence</td>\n",
       "      <td>\"Suboxone has completely turned my life around...</td>\n",
       "      <td>9</td>\n",
       "      <td>27-Nov-16</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uniqueID                  drugName                     condition  \\\n",
       "0    206461                 Valsartan  Left Ventricular Dysfunction   \n",
       "1     95260                Guanfacine                          ADHD   \n",
       "2     92703                    Lybrel                 Birth Control   \n",
       "3    138000                Ortho Evra                 Birth Control   \n",
       "4     35696  Buprenorphine / naloxone             Opiate Dependence   \n",
       "\n",
       "                                              review  rating       date  \\\n",
       "0  \"It has no side effect, I take it in combinati...       9  20-May-12   \n",
       "1  \"My son is halfway through his fourth week of ...       8  27-Apr-10   \n",
       "2  \"I used to take another oral contraceptive, wh...       5  14-Dec-09   \n",
       "3  \"This is my first time using any form of birth...       8   3-Nov-15   \n",
       "4  \"Suboxone has completely turned my life around...       9  27-Nov-16   \n",
       "\n",
       "   usefulCount  \n",
       "0           27  \n",
       "1          192  \n",
       "2           17  \n",
       "3           10  \n",
       "4           37  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Combine Two Dataset - Train and Test Dataset\n",
    "data = pd.concat([train, test])\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b76d294f-16c2-4f95-9881-04f4a484d352",
   "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>uniqueID</th>\n",
       "      <th>rating</th>\n",
       "      <th>usefulCount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>215063.000000</td>\n",
       "      <td>215063.000000</td>\n",
       "      <td>215063.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>116039.364814</td>\n",
       "      <td>6.990008</td>\n",
       "      <td>28.001004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>67007.913366</td>\n",
       "      <td>3.275554</td>\n",
       "      <td>36.346069</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>58115.500000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>6.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>115867.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>16.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>173963.500000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>36.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>232291.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>1291.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            uniqueID         rating    usefulCount\n",
       "count  215063.000000  215063.000000  215063.000000\n",
       "mean   116039.364814       6.990008      28.001004\n",
       "std     67007.913366       3.275554      36.346069\n",
       "min         0.000000       1.000000       0.000000\n",
       "25%     58115.500000       5.000000       6.000000\n",
       "50%    115867.000000       8.000000      16.000000\n",
       "75%    173963.500000      10.000000      36.000000\n",
       "max    232291.000000      10.000000    1291.000000"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5c640707-62be-4edf-a83f-32eb53d55abf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Index: 215063 entries, 0 to 53765\n",
      "Data columns (total 7 columns):\n",
      " #   Column       Non-Null Count   Dtype \n",
      "---  ------       --------------   ----- \n",
      " 0   uniqueID     215063 non-null  int64 \n",
      " 1   drugName     215063 non-null  object\n",
      " 2   condition    213869 non-null  object\n",
      " 3   review       215063 non-null  object\n",
      " 4   rating       215063 non-null  int64 \n",
      " 5   date         215063 non-null  object\n",
      " 6   usefulCount  215063 non-null  int64 \n",
      "dtypes: int64(3), object(4)\n",
      "memory usage: 13.1+ MB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "4d5a8667-3a83-4bc6-99b9-3d132b239d9f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "uniqueID        int64\n",
       "drugName       object\n",
       "condition      object\n",
       "review         object\n",
       "rating          int64\n",
       "date           object\n",
       "usefulCount     int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "4401403b-38d1-45e6-a19d-8a974d3b48fe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "uniqueID       False\n",
       "drugName       False\n",
       "condition       True\n",
       "review         False\n",
       "rating         False\n",
       "date           False\n",
       "usefulCount    False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isnull().any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "1c74663b-b727-46c1-8201-ca86735772dd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Review_Sentiment\n",
       "1.0    150768\n",
       "0.0     64295\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Creating a new column Review_Sentiment\n",
    "data.loc[(data['rating'] > 5), 'Review_Sentiment'] = 1\n",
    "data.loc[(data['rating'] <= 5), 'Review_Sentiment'] = 0\n",
    "\n",
    "data['Review_Sentiment'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "16d91a1c-6479-4c0d-947e-ca83361a0c66",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Converting the date into datetime format\n",
    "data['date'] = pd.to_datetime(data['date'], errors = 'coerce')\n",
    "\n",
    "# Extracting year, month, date from date\n",
    "data['Year'] = data['date'].dt.year\n",
    "data['Month'] = data['date'].dt.month\n",
    "data['Day'] = data['date'].dt.day"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d62638dd-262e-4986-bcc7-49479c2ca065",
   "metadata": {},
   "source": [
    "### Data Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "7a1167f9-2b3d-4aa3-99b4-077aaccb1535",
   "metadata": {},
   "outputs": [],
   "source": [
    "def review_clean(review):\n",
    "    # Changing reviews to lower case\n",
    "    to_lower = review.str.lower()\n",
    "    \n",
    "    # Replacing the repeating pattern of &#039;\n",
    "    repeat_pattern_remove = to_lower.str.replace(\"&#039;\", \"\")\n",
    "    \n",
    "    # Removing all special characters\n",
    "    special_char_remove = repeat_pattern_remove.str.replace(r'[^\\w\\d\\s]',' ')\n",
    "    \n",
    "    # Removing all ASCII characters\n",
    "    non_ascii_char_remove = special_char_remove.str.replace(r'[^\\x00-\\x7F]+',' ')\n",
    "    \n",
    "    # Removing the leading and trailing whitespaces\n",
    "    whitespace_remove = non_ascii_char_remove.str.replace(r'^\\s+|\\s+?$','')\n",
    "    \n",
    "    # Replacing multiple spaces with single space\n",
    "    multiple_space_remove = whitespace_remove.str.replace(r'\\s+',' ')\n",
    "    \n",
    "    # Replacing two or more dots with one\n",
    "    dataframe = multiple_space_remove.str.replace(r'\\.{2,}', ' ')\n",
    "\n",
    "    return dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "b73d7e61-0b9f-4373-b3a0-fc444afff3b7",
   "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>uniqueID</th>\n",
       "      <th>drugName</th>\n",
       "      <th>condition</th>\n",
       "      <th>review</th>\n",
       "      <th>rating</th>\n",
       "      <th>date</th>\n",
       "      <th>usefulCount</th>\n",
       "      <th>Review_Sentiment</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Day</th>\n",
       "      <th>review_clean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>206461</td>\n",
       "      <td>Valsartan</td>\n",
       "      <td>Left Ventricular Dysfunction</td>\n",
       "      <td>\"It has no side effect, I take it in combinati...</td>\n",
       "      <td>9</td>\n",
       "      <td>2012-05-20</td>\n",
       "      <td>27</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2012</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>\"it has no side effect, i take it in combinati...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>95260</td>\n",
       "      <td>Guanfacine</td>\n",
       "      <td>ADHD</td>\n",
       "      <td>\"My son is halfway through his fourth week of ...</td>\n",
       "      <td>8</td>\n",
       "      <td>2010-04-27</td>\n",
       "      <td>192</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2010</td>\n",
       "      <td>4</td>\n",
       "      <td>27</td>\n",
       "      <td>\"my son is halfway through his fourth week of ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>92703</td>\n",
       "      <td>Lybrel</td>\n",
       "      <td>Birth Control</td>\n",
       "      <td>\"I used to take another oral contraceptive, wh...</td>\n",
       "      <td>5</td>\n",
       "      <td>2009-12-14</td>\n",
       "      <td>17</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2009</td>\n",
       "      <td>12</td>\n",
       "      <td>14</td>\n",
       "      <td>\"i used to take another oral contraceptive, wh...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uniqueID    drugName                     condition  \\\n",
       "0    206461   Valsartan  Left Ventricular Dysfunction   \n",
       "1     95260  Guanfacine                          ADHD   \n",
       "2     92703      Lybrel                 Birth Control   \n",
       "\n",
       "                                              review  rating       date  \\\n",
       "0  \"It has no side effect, I take it in combinati...       9 2012-05-20   \n",
       "1  \"My son is halfway through his fourth week of ...       8 2010-04-27   \n",
       "2  \"I used to take another oral contraceptive, wh...       5 2009-12-14   \n",
       "\n",
       "   usefulCount  Review_Sentiment  Year  Month  Day  \\\n",
       "0           27               1.0  2012      5   20   \n",
       "1          192               1.0  2010      4   27   \n",
       "2           17               0.0  2009     12   14   \n",
       "\n",
       "                                        review_clean  \n",
       "0  \"it has no side effect, i take it in combinati...  \n",
       "1  \"my son is halfway through his fourth week of ...  \n",
       "2  \"i used to take another oral contraceptive, wh...  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['review_clean'] = review_clean(data['review'])\n",
    "data.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "f976495c-ad32-4e81-96d6-fe1991ec7112",
   "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>uniqueID</th>\n",
       "      <th>drugName</th>\n",
       "      <th>condition</th>\n",
       "      <th>review</th>\n",
       "      <th>rating</th>\n",
       "      <th>date</th>\n",
       "      <th>usefulCount</th>\n",
       "      <th>Review_Sentiment</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Day</th>\n",
       "      <th>review_clean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>206461</td>\n",
       "      <td>Valsartan</td>\n",
       "      <td>Left Ventricular Dysfunction</td>\n",
       "      <td>\"It has no side effect, I take it in combinati...</td>\n",
       "      <td>9</td>\n",
       "      <td>2012-05-20</td>\n",
       "      <td>27</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2012</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>\"it has no side effect, i take it in combinati...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>95260</td>\n",
       "      <td>Guanfacine</td>\n",
       "      <td>ADHD</td>\n",
       "      <td>\"My son is halfway through his fourth week of ...</td>\n",
       "      <td>8</td>\n",
       "      <td>2010-04-27</td>\n",
       "      <td>192</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2010</td>\n",
       "      <td>4</td>\n",
       "      <td>27</td>\n",
       "      <td>\"my son is halfway through his fourth week of ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>92703</td>\n",
       "      <td>Lybrel</td>\n",
       "      <td>Birth Control</td>\n",
       "      <td>\"I used to take another oral contraceptive, wh...</td>\n",
       "      <td>5</td>\n",
       "      <td>2009-12-14</td>\n",
       "      <td>17</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2009</td>\n",
       "      <td>12</td>\n",
       "      <td>14</td>\n",
       "      <td>\"i used to take another oral contraceptive, wh...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uniqueID    drugName                     condition  \\\n",
       "0    206461   Valsartan  Left Ventricular Dysfunction   \n",
       "1     95260  Guanfacine                          ADHD   \n",
       "2     92703      Lybrel                 Birth Control   \n",
       "\n",
       "                                              review  rating       date  \\\n",
       "0  \"It has no side effect, I take it in combinati...       9 2012-05-20   \n",
       "1  \"My son is halfway through his fourth week of ...       8 2010-04-27   \n",
       "2  \"I used to take another oral contraceptive, wh...       5 2009-12-14   \n",
       "\n",
       "   usefulCount  Review_Sentiment  Year  Month  Day  \\\n",
       "0           27               1.0  2012      5   20   \n",
       "1          192               1.0  2010      4   27   \n",
       "2           17               0.0  2009     12   14   \n",
       "\n",
       "                                        review_clean  \n",
       "0  \"it has no side effect, i take it in combinati...  \n",
       "1  \"my son is halfway through his fourth week of ...  \n",
       "2  \"i used to take another oral contraceptive, wh...  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Contraction Dictionary for the expansion\n",
    "import re\n",
    "contractions_dict = {\n",
    "    \"ain't\": \"am not\", \"aren't\": \"are not\", \"can't\": \"cannot\", \"can't've\": \"cannot have\", \"'cause\": \"because\",\n",
    "    \"could've\": \"could have\", \"couldn't\": \"could not\", \"couldn't've\": \"could not have\", \"didn't\": \"did not\", \"doesn't\": \"does not\",\n",
    "    \"doesn’t\": \"does not\", \"don't\": \"do not\", \"don’t\": \"do not\", \"hadn't\": \"had not\", \"hadn't've\": \"had not have\", \"hasn't\": \"has not\",\n",
    "    \"haven't\": \"have not\", \"he'd\": \"he had\", \"he'd've\": \"he would have\", \"he'll\": \"he will\", \"he'll've\": \"he will have\", \"he's\": \"he is\",\n",
    "    \"how'd\": \"how did\", \"how'd'y\": \"how do you\", \"how'll\": \"how will\", \"how's\": \"how is\", \"i'd\": \"i would\", \"i'd've\": \"i would have\",\n",
    "    \"i'll\": \"i will\", \"i'll've\": \"i will have\", \"i'm\": \"i am\", \"i've\": \"i have\", \"isn't\": \"is not\", \"it'd\": \"it would\", \"it'd've\": \"it would have\",\n",
    "    \"it'll\": \"it will\", \"it'll've\": \"it will have\", \"it's\": \"it is\", \"let's\": \"let us\", \"ma'am\": \"madam\", \"mayn't\": \"may not\",\"might've\": \"might have\",\n",
    "    \"mightn't\": \"might not\", \"mightn't've\": \"might not have\", \"must've\": \"must have\", \"mustn't\": \"must not\", \"mustn't've\": \"must not have\",\n",
    "    \"needn't\": \"need not\", \"needn't've\": \"need not have\", \"o'clock\": \"of the clock\", \"oughtn't\": \"ought not\", \"oughtn't've\": \"ought not have\",\n",
    "    \"shan't\": \"shall not\",\"sha'n't\": \"shall not\", \"shan't've\": \"shall not have\", \"she'd\": \"she would\", \"she'd've\": \"she would have\",\n",
    "    \"she'll\": \"she will\", \"she'll've\": \"she will have\", \"she's\": \"she is\", \"should've\": \"should have\", \"shouldn't\": \"should not\", \n",
    "    \"shouldn't've\": \"should not have\", \"so've\": \"so have\", \"so's\": \"so is\", \"that'd\": \"that would\", \"that'd've\": \"that would have\",\n",
    "    \"that's\": \"that is\", \"there'd\": \"there would\", \"there'd've\": \"there would have\", \"there's\": \"there is\", \"they'd\": \"they would\",\n",
    "    \"they'd've\": \"they would have\", \"they'll\": \"they will\", \"they'll've\": \"they will have\", \"they're\": \"they are\", \"they've\": \"they have\",\n",
    "    \"to've\": \"to have\", \"wasn't\": \"was not\", \"we'd\": \"we would\", \"we'd've\": \"we would have\", \"we'll\": \"we will\", \"we'll've\": \"we will have\",\n",
    "    \"we're\": \"we are\", \"we've\": \"we have\", \"weren't\": \"were not\", \"what'll\": \"what will\", \"what'll've\": \"what will have\", \"what're\": \"what are\",\n",
    "    \"what's\": \"what is\", \"what've\": \"what have\", \"when's\": \"when is\", \"when've\": \"when have\", \"where'd\": \"where did\", \"where's\": \"where is\",\n",
    "    \"where've\": \"where have\", \"who'll\": \"who will\", \"who'll've\": \"who will have\", \"who's\": \"who is\", \"who've\": \"who have\", \"why's\": \"why is\",\n",
    "    \"why've\": \"why have\", \"will've\": \"will have\", \"won't\": \"will not\", \"won't've\": \"will not have\", \"would've\": \"would have\",\n",
    "    \"wouldn't\": \"would not\", \"wouldn't've\": \"would not have\", \"y'all\": \"you all\", \"y’all\": \"you all\", \"y'all'd\": \"you all would\",\n",
    "    \"y'all'd've\": \"you all would have\", \"y'all're\": \"you all are\", \"y'all've\": \"you all have\", \"you'd\": \"you would\", \"you'd've\": \"you would have\",\n",
    "    \"you'll\": \"you will\", \"you'll've\": \"you will have\", \"you're\": \"you are\", \"you've\": \"you have\", \"ain’t\": \"am not\", \"aren’t\": \"are not\",\n",
    "    \"can’t\": \"cannot\", \"can’t’ve\": \"cannot have\", \"’cause\": \"because\", \"could’ve\": \"could have\", \"couldn’t\": \"could not\", \"couldn’t’ve\": \"could not have\",\n",
    "    \"didn’t\": \"did not\", \"doesn’t\": \"does not\", \"don’t\": \"do not\", \"don’t\": \"do not\", \"hadn’t\": \"had not\", \"hadn’t’ve\": \"had not have\",\n",
    "    \"hasn’t\": \"has not\", \"haven’t\": \"have not\", \"he’d\": \"he had\", \"he’d’ve\": \"he would have\", \"he’ll\": \"he will\", \"he’ll’ve\": \"he will have\",\n",
    "    \"he’s\": \"he is\", \"how’d\": \"how did\", \"how’d’y\": \"how do you\", \"how’ll\": \"how will\", \"how’s\": \"how is\", \"i’d\": \"i would\", \"i’d’ve\": \"i would have\",\n",
    "    \"i’ll\": \"i will\", \"i’ll’ve\": \"i will have\", \"i’m\": \"i am\", \"i’ve\": \"i have\", \"isn’t\": \"is not\", \"it’d\": \"it would\", \"it’d’ve\": \"it would have\",\n",
    "    \"it’ll\": \"it will\", \"it’ll’ve\": \"it will have\", \"it’s\": \"it is\", \"let’s\": \"let us\", \"ma’am\": \"madam\", \"mayn’t\": \"may not\",\n",
    "    \"might’ve\": \"might have\", \"mightn’t\": \"might not\", \"mightn’t’ve\": \"might not have\", \"must’ve\": \"must have\", \"mustn’t\": \"must not\",\n",
    "    \"mustn’t’ve\": \"must not have\", \"needn’t\": \"need not\", \"needn’t’ve\": \"need not have\", \"o’clock\": \"of the clock\",\n",
    "    \"oughtn’t\": \"ought not\", \"oughtn’t’ve\": \"ought not have\", \"shan’t\": \"shall not\", \"sha’n’t\": \"shall not\", \"shan’t’ve\": \"shall not have\",\n",
    "    \"she’d\": \"she would\", \"she’d’ve\": \"she would have\", \"she’ll\": \"she will\", \"she’ll’ve\": \"she will have\", \"she’s\": \"she is\",\n",
    "    \"should’ve\": \"should have\", \"shouldn’t\": \"should not\", \"shouldn’t’ve\": \"should not have\", \"so’ve\": \"so have\", \"so’s\": \"so is\",\n",
    "    \"that’d\": \"that would\", \"that’d’ve\": \"that would have\", \"that’s\": \"that is\", \"there’d\": \"there would\", \"there’d’ve\": \"there would have\",\n",
    "    \"there’s\": \"there is\", \"they’d\": \"they would\", \"they’d’ve\": \"they would have\", \"they’ll\": \"they will\", \"they’ll’ve\": \"they will have\",\n",
    "    \"they’re\": \"they are\", \"they’ve\": \"they have\", \"to’ve\": \"to have\", \"wasn’t\": \"was not\", \"we’d\": \"we would\", \"we’d’ve\": \"we would have\",\n",
    "    \"we’ll\": \"we will\", \"we’ll’ve\": \"we will have\", \"we’re\": \"we are\", \"we’ve\": \"we have\", \"weren’t\": \"were not\", \"what’ll\": \"what will\",\n",
    "    \"what’ll’ve\": \"what will have\", \"what’re\": \"what are\", \"what’s\": \"what is\", \"what’ve\": \"what have\", \"when’s\": \"when is\",\n",
    "    \"when’ve\": \"when have\", \"where’d\": \"where did\", \"where’s\": \"where is\", \"where’ve\": \"where have\", \"who’ll\": \"who will\",\n",
    "    \"who’ll’ve\": \"who will have\", \"who’s\": \"who is\", \"who’ve\": \"who have\",\"why’s\": \"why is\", \"why’ve\": \"why have\", \"will’ve\": \"will have\",\n",
    "    \"won’t\": \"will not\", \"won’t’ve\": \"will not have\", \"would’ve\": \"would have\", \"wouldn’t\": \"would not\", \"wouldn’t’ve\": \"would not have\",\n",
    "    \"y’all\": \"you all\", \"y’all\": \"you all\", \"y’all’d\": \"you all would\", \"y’all’d’ve\": \"you all would have\", \"y’all’re\": \"you all are\",\n",
    "    \"y’all’ve\": \"you all have\", \"you’d\": \"you would\", \"you’d’ve\": \"you would have\", \"you’ll\": \"you will\", \"you’ll’ve\": \"you will have\",\n",
    "    \"you’re\": \"you are\", \"you’re\": \"you are\", \"you’ve\": \"you have\"\n",
    "}\n",
    "contractions_re = re.compile('(%s)' % '|'.join(contractions_dict.keys()))\n",
    "\n",
    "# Function expand the contractions if there's any\n",
    "def expand_contractions(s, contractions_dict=contractions_dict):\n",
    "    def replace(match):\n",
    "        return contractions_dict[match.group(0)]\n",
    "    return contractions_re.sub(replace, s)\n",
    "\n",
    "\n",
    "# Expanding the contractions\n",
    "data['review_clean'] = data['review_clean'].apply(lambda x: expand_contractions(x))\n",
    "data.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "b4246878-9511-4c53-b506-53d16671d094",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Using Stopwords to extract the useful words in review_clean\n",
    "import nltk\n",
    "from nltk.corpus import stopwords\n",
    "\n",
    "stop_words = set(stopwords.words('english'))\n",
    "data['review_clean'] = data['review_clean'].apply(lambda x: ' '.join(word for word in x.split() if word not in stop_words))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "a44e71e1-012e-47d7-8d0c-9a3aa2235e80",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Removing the word stems using the Snowball Stemmer in review_clean\n",
    "import nltk\n",
    "from nltk.stem import SnowballStemmer\n",
    "\n",
    "snow_ball_stemmer = SnowballStemmer(language = \"english\")\n",
    "data['review_clean'] = data['review_clean'].apply(lambda x: \" \".join(snow_ball_stemmer.stem(word) for word in x.split()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "e47b4e6e-8d71-4441-92c1-08bfdeb1a769",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Sentiment polarity of all reviews\n",
    "from textblob import TextBlob\n",
    "\n",
    "def sentiment_polarity(review):\n",
    "    pol = []\n",
    "    for i in review:\n",
    "        analysis = TextBlob(i)\n",
    "        pol.append(analysis.sentiment.polarity)\n",
    "    return pol"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e0f5ada5-e818-4b1e-b1de-cc1baee8ed2d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Evaluating Sentiment Polarity for Uncleaned Reviews\n",
    "data['sentiment'] = sentiment_polarity(data['review'])\n",
    "\n",
    "# Evaluating Sentiment Polarity for Cleaned Reviews\n",
    "data['sentiment_clean'] = sentiment_polarity(data['review_clean'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "405d326a-3c01-433e-9c2c-affa43bacb38",
   "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>uniqueID</th>\n",
       "      <th>drugName</th>\n",
       "      <th>condition</th>\n",
       "      <th>review</th>\n",
       "      <th>rating</th>\n",
       "      <th>date</th>\n",
       "      <th>usefulCount</th>\n",
       "      <th>Review_Sentiment</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>Day</th>\n",
       "      <th>review_clean</th>\n",
       "      <th>sentiment</th>\n",
       "      <th>sentiment_clean</th>\n",
       "      <th>review_clean_ss</th>\n",
       "      <th>sentiment_clean_ss</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>206461</td>\n",
       "      <td>Valsartan</td>\n",
       "      <td>Left Ventricular Dysfunction</td>\n",
       "      <td>\"It has no side effect, I take it in combinati...</td>\n",
       "      <td>9</td>\n",
       "      <td>2012-05-20</td>\n",
       "      <td>27</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2012</td>\n",
       "      <td>5</td>\n",
       "      <td>20</td>\n",
       "      <td>\"it side effect, take combin bystol 5 mg fish ...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>\"it has no side effect, i take it in combinati...</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>95260</td>\n",
       "      <td>Guanfacine</td>\n",
       "      <td>ADHD</td>\n",
       "      <td>\"My son is halfway through his fourth week of ...</td>\n",
       "      <td>8</td>\n",
       "      <td>2010-04-27</td>\n",
       "      <td>192</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2010</td>\n",
       "      <td>4</td>\n",
       "      <td>27</td>\n",
       "      <td>\"mi son halfway fourth week intuniv. becam con...</td>\n",
       "      <td>0.168333</td>\n",
       "      <td>0.151923</td>\n",
       "      <td>\"my son is halfway through his fourth week of ...</td>\n",
       "      <td>0.168333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>92703</td>\n",
       "      <td>Lybrel</td>\n",
       "      <td>Birth Control</td>\n",
       "      <td>\"I used to take another oral contraceptive, wh...</td>\n",
       "      <td>5</td>\n",
       "      <td>2009-12-14</td>\n",
       "      <td>17</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2009</td>\n",
       "      <td>12</td>\n",
       "      <td>14</td>\n",
       "      <td>\"i use take anoth oral contraceptive, 21 pill ...</td>\n",
       "      <td>0.067210</td>\n",
       "      <td>0.080769</td>\n",
       "      <td>\"i used to take another oral contraceptive, wh...</td>\n",
       "      <td>0.067210</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   uniqueID    drugName                     condition  \\\n",
       "0    206461   Valsartan  Left Ventricular Dysfunction   \n",
       "1     95260  Guanfacine                          ADHD   \n",
       "2     92703      Lybrel                 Birth Control   \n",
       "\n",
       "                                              review  rating       date  \\\n",
       "0  \"It has no side effect, I take it in combinati...       9 2012-05-20   \n",
       "1  \"My son is halfway through his fourth week of ...       8 2010-04-27   \n",
       "2  \"I used to take another oral contraceptive, wh...       5 2009-12-14   \n",
       "\n",
       "   usefulCount  Review_Sentiment  Year  Month  Day  \\\n",
       "0           27               1.0  2012      5   20   \n",
       "1          192               1.0  2010      4   27   \n",
       "2           17               0.0  2009     12   14   \n",
       "\n",
       "                                        review_clean  sentiment  \\\n",
       "0  \"it side effect, take combin bystol 5 mg fish ...   0.000000   \n",
       "1  \"mi son halfway fourth week intuniv. becam con...   0.168333   \n",
       "2  \"i use take anoth oral contraceptive, 21 pill ...   0.067210   \n",
       "\n",
       "   sentiment_clean                                    review_clean_ss  \\\n",
       "0         0.000000  \"it has no side effect, i take it in combinati...   \n",
       "1         0.151923  \"my son is halfway through his fourth week of ...   \n",
       "2         0.080769  \"i used to take another oral contraceptive, wh...   \n",
       "\n",
       "   sentiment_clean_ss  \n",
       "0            0.000000  \n",
       "1            0.168333  \n",
       "2            0.067210  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Cleaning the reviews without using Snowball Stemmer and Stopwords\n",
    "data['review_clean_ss'] = review_clean(data['review'])\n",
    "\n",
    "# Evaluating Sentiment Polarity for cleaned Reviews without using Snowball Stemmer and Stopwords\n",
    "data['sentiment_clean_ss'] = sentiment_polarity(data['review_clean_ss'])\n",
    "data.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "1b5ba1f5-9060-4c72-959b-8e07e5acdf7b",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Remove Missing Values\n",
    "data = data.dropna(how=\"any\", axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "9ea116d1-021d-4aa9-ba53-277c4798a879",
   "metadata": {},
   "outputs": [],
   "source": [
    "import string\n",
    "\n",
    "# Words count in each review\n",
    "data['words_count']=data[\"review_clean_ss\"].apply(lambda x: len(str(x).split()))\n",
    "\n",
    "# Unique words count in each review\n",
    "data['unique_words_count']=data[\"review_clean_ss\"].apply(lambda x: len(set(str(x).split())))\n",
    "\n",
    "# Letters count in each review\n",
    "data['letters_count']=data[\"review_clean_ss\"].apply(lambda x: len(str(x)))\n",
    "\n",
    "# Punctuations count in each review\n",
    "data[\"punctuations_count\"] = data[\"review\"].apply(lambda x: len([c for c in str(x) if c in string.punctuation]))\n",
    "\n",
    "# Upper case words count in each review\n",
    "data[\"uppercase_words_count\"] = data[\"review\"].apply(lambda x: len([w for w in str(x).split() if w.isupper()]))\n",
    "\n",
    "# Title case words count in each review\n",
    "data[\"titlecase_words_count\"] = data[\"review\"].apply(lambda x: len([w for w in str(x).split() if w.istitle()]))\n",
    "\n",
    "# Stopwords count in each review\n",
    "data[\"stopwords_count\"] = data[\"review\"].apply(lambda x: len([w for w in str(x).lower().split() if w in stop_words]))\n",
    "\n",
    "# Average length of the words in each review\n",
    "data[\"average_word_length\"] = data[\"review_clean_ss\"].apply(lambda x: np.mean([len(w) for w in str(x).split()]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "6a3ef24d-0762-4380-a838-ca4c9220b661",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['uniqueID', 'drugName', 'condition', 'review', 'rating', 'date',\n",
       "       'usefulCount', 'Review_Sentiment', 'Year', 'Month', 'Day',\n",
       "       'review_clean', 'sentiment', 'sentiment_clean', 'review_clean_ss',\n",
       "       'sentiment_clean_ss', 'words_count', 'unique_words_count',\n",
       "       'letters_count', 'punctuations_count', 'uppercase_words_count',\n",
       "       'titlecase_words_count', 'stopwords_count', 'average_word_length'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "1711b781-00c2-4cc2-aaf3-6ecf78c26c65",
   "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>uniqueID</th>\n",
       "      <th>drugName</th>\n",
       "      <th>condition</th>\n",
       "      <th>review</th>\n",
       "      <th>rating</th>\n",
       "      <th>date</th>\n",
       "      <th>usefulCount</th>\n",
       "      <th>Review_Sentiment</th>\n",
       "      <th>Year</th>\n",
       "      <th>Month</th>\n",
       "      <th>...</th>\n",
       "      <th>review_clean_ss</th>\n",
       "      <th>sentiment_clean_ss</th>\n",
       "      <th>words_count</th>\n",
       "      <th>unique_words_count</th>\n",
       "      <th>letters_count</th>\n",
       "      <th>punctuations_count</th>\n",
       "      <th>uppercase_words_count</th>\n",
       "      <th>titlecase_words_count</th>\n",
       "      <th>stopwords_count</th>\n",
       "      <th>average_word_length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>206461</td>\n",
       "      <td>3421</td>\n",
       "      <td>480</td>\n",
       "      <td>\"It has no side effect, I take it in combinati...</td>\n",
       "      <td>9</td>\n",
       "      <td>2012-05-20</td>\n",
       "      <td>27</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2012</td>\n",
       "      <td>5</td>\n",
       "      <td>...</td>\n",
       "      <td>\"it has no side effect, i take it in combinati...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>17</td>\n",
       "      <td>17</td>\n",
       "      <td>79</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>7</td>\n",
       "      <td>3.705882</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>95260</td>\n",
       "      <td>1536</td>\n",
       "      <td>80</td>\n",
       "      <td>\"My son is halfway through his fourth week of ...</td>\n",
       "      <td>8</td>\n",
       "      <td>2010-04-27</td>\n",
       "      <td>192</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2010</td>\n",
       "      <td>4</td>\n",
       "      <td>...</td>\n",
       "      <td>\"my son is halfway through his fourth week of ...</td>\n",
       "      <td>0.168333</td>\n",
       "      <td>141</td>\n",
       "      <td>106</td>\n",
       "      <td>741</td>\n",
       "      <td>23</td>\n",
       "      <td>2</td>\n",
       "      <td>13</td>\n",
       "      <td>69</td>\n",
       "      <td>4.248227</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>92703</td>\n",
       "      <td>1983</td>\n",
       "      <td>175</td>\n",
       "      <td>\"I used to take another oral contraceptive, wh...</td>\n",
       "      <td>5</td>\n",
       "      <td>2009-12-14</td>\n",
       "      <td>17</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2009</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>\"i used to take another oral contraceptive, wh...</td>\n",
       "      <td>0.067210</td>\n",
       "      <td>134</td>\n",
       "      <td>95</td>\n",
       "      <td>743</td>\n",
       "      <td>34</td>\n",
       "      <td>6</td>\n",
       "      <td>15</td>\n",
       "      <td>59</td>\n",
       "      <td>4.544776</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>138000</td>\n",
       "      <td>2446</td>\n",
       "      <td>175</td>\n",
       "      <td>\"This is my first time using any form of birth...</td>\n",
       "      <td>8</td>\n",
       "      <td>2015-11-03</td>\n",
       "      <td>10</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2015</td>\n",
       "      <td>11</td>\n",
       "      <td>...</td>\n",
       "      <td>\"this is my first time using any form of birth...</td>\n",
       "      <td>0.179545</td>\n",
       "      <td>89</td>\n",
       "      <td>57</td>\n",
       "      <td>442</td>\n",
       "      <td>15</td>\n",
       "      <td>4</td>\n",
       "      <td>9</td>\n",
       "      <td>45</td>\n",
       "      <td>3.977528</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>35696</td>\n",
       "      <td>553</td>\n",
       "      <td>593</td>\n",
       "      <td>\"Suboxone has completely turned my life around...</td>\n",
       "      <td>9</td>\n",
       "      <td>2016-11-27</td>\n",
       "      <td>37</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2016</td>\n",
       "      <td>11</td>\n",
       "      <td>...</td>\n",
       "      <td>\"suboxone has completely turned my life around...</td>\n",
       "      <td>0.194444</td>\n",
       "      <td>124</td>\n",
       "      <td>86</td>\n",
       "      <td>695</td>\n",
       "      <td>28</td>\n",
       "      <td>7</td>\n",
       "      <td>15</td>\n",
       "      <td>60</td>\n",
       "      <td>4.532258</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>155963</td>\n",
       "      <td>715</td>\n",
       "      <td>170</td>\n",
       "      <td>\"2nd day on 5mg started to work with rock hard...</td>\n",
       "      <td>2</td>\n",
       "      <td>2015-11-28</td>\n",
       "      <td>43</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2015</td>\n",
       "      <td>11</td>\n",
       "      <td>...</td>\n",
       "      <td>\"2nd day on 5mg started to work with rock hard...</td>\n",
       "      <td>0.003472</td>\n",
       "      <td>68</td>\n",
       "      <td>57</td>\n",
       "      <td>391</td>\n",
       "      <td>22</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>18</td>\n",
       "      <td>4.735294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>165907</td>\n",
       "      <td>1874</td>\n",
       "      <td>314</td>\n",
       "      <td>\"He pulled out, but he cummed a bit in me. I t...</td>\n",
       "      <td>1</td>\n",
       "      <td>2017-03-07</td>\n",
       "      <td>5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>\"he pulled out, but he cummed a bit in me. i t...</td>\n",
       "      <td>0.111111</td>\n",
       "      <td>30</td>\n",
       "      <td>27</td>\n",
       "      <td>136</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>3.566667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>102654</td>\n",
       "      <td>284</td>\n",
       "      <td>174</td>\n",
       "      <td>\"Abilify changed my life. There is hope. I was...</td>\n",
       "      <td>10</td>\n",
       "      <td>2015-03-14</td>\n",
       "      <td>32</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2015</td>\n",
       "      <td>3</td>\n",
       "      <td>...</td>\n",
       "      <td>\"abilify changed my life. there is hope. i was...</td>\n",
       "      <td>0.074107</td>\n",
       "      <td>146</td>\n",
       "      <td>94</td>\n",
       "      <td>727</td>\n",
       "      <td>24</td>\n",
       "      <td>13</td>\n",
       "      <td>27</td>\n",
       "      <td>77</td>\n",
       "      <td>3.979452</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>74811</td>\n",
       "      <td>1779</td>\n",
       "      <td>324</td>\n",
       "      <td>\" I Ve had  nothing but problems with the Kepp...</td>\n",
       "      <td>1</td>\n",
       "      <td>2016-08-09</td>\n",
       "      <td>11</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2016</td>\n",
       "      <td>8</td>\n",
       "      <td>...</td>\n",
       "      <td>\" i ve had  nothing but problems with the kepp...</td>\n",
       "      <td>0.200000</td>\n",
       "      <td>35</td>\n",
       "      <td>27</td>\n",
       "      <td>193</td>\n",
       "      <td>14</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>11</td>\n",
       "      <td>4.514286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>48928</td>\n",
       "      <td>1263</td>\n",
       "      <td>175</td>\n",
       "      <td>\"I had been on the pill for many years. When m...</td>\n",
       "      <td>8</td>\n",
       "      <td>2016-12-08</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2016</td>\n",
       "      <td>12</td>\n",
       "      <td>...</td>\n",
       "      <td>\"i had been on the pill for many years. when m...</td>\n",
       "      <td>0.079167</td>\n",
       "      <td>142</td>\n",
       "      <td>102</td>\n",
       "      <td>719</td>\n",
       "      <td>28</td>\n",
       "      <td>13</td>\n",
       "      <td>16</td>\n",
       "      <td>66</td>\n",
       "      <td>4.070423</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   uniqueID  drugName  condition  \\\n",
       "0    206461      3421        480   \n",
       "1     95260      1536         80   \n",
       "2     92703      1983        175   \n",
       "3    138000      2446        175   \n",
       "4     35696       553        593   \n",
       "5    155963       715        170   \n",
       "6    165907      1874        314   \n",
       "7    102654       284        174   \n",
       "8     74811      1779        324   \n",
       "9     48928      1263        175   \n",
       "\n",
       "                                              review  rating       date  \\\n",
       "0  \"It has no side effect, I take it in combinati...       9 2012-05-20   \n",
       "1  \"My son is halfway through his fourth week of ...       8 2010-04-27   \n",
       "2  \"I used to take another oral contraceptive, wh...       5 2009-12-14   \n",
       "3  \"This is my first time using any form of birth...       8 2015-11-03   \n",
       "4  \"Suboxone has completely turned my life around...       9 2016-11-27   \n",
       "5  \"2nd day on 5mg started to work with rock hard...       2 2015-11-28   \n",
       "6  \"He pulled out, but he cummed a bit in me. I t...       1 2017-03-07   \n",
       "7  \"Abilify changed my life. There is hope. I was...      10 2015-03-14   \n",
       "8  \" I Ve had  nothing but problems with the Kepp...       1 2016-08-09   \n",
       "9  \"I had been on the pill for many years. When m...       8 2016-12-08   \n",
       "\n",
       "   usefulCount  Review_Sentiment  Year  Month  ...  \\\n",
       "0           27               1.0  2012      5  ...   \n",
       "1          192               1.0  2010      4  ...   \n",
       "2           17               0.0  2009     12  ...   \n",
       "3           10               1.0  2015     11  ...   \n",
       "4           37               1.0  2016     11  ...   \n",
       "5           43               0.0  2015     11  ...   \n",
       "6            5               0.0  2017      3  ...   \n",
       "7           32               1.0  2015      3  ...   \n",
       "8           11               0.0  2016      8  ...   \n",
       "9            1               1.0  2016     12  ...   \n",
       "\n",
       "                                     review_clean_ss sentiment_clean_ss  \\\n",
       "0  \"it has no side effect, i take it in combinati...           0.000000   \n",
       "1  \"my son is halfway through his fourth week of ...           0.168333   \n",
       "2  \"i used to take another oral contraceptive, wh...           0.067210   \n",
       "3  \"this is my first time using any form of birth...           0.179545   \n",
       "4  \"suboxone has completely turned my life around...           0.194444   \n",
       "5  \"2nd day on 5mg started to work with rock hard...           0.003472   \n",
       "6  \"he pulled out, but he cummed a bit in me. i t...           0.111111   \n",
       "7  \"abilify changed my life. there is hope. i was...           0.074107   \n",
       "8  \" i ve had  nothing but problems with the kepp...           0.200000   \n",
       "9  \"i had been on the pill for many years. when m...           0.079167   \n",
       "\n",
       "   words_count  unique_words_count letters_count  punctuations_count  \\\n",
       "0           17                  17            79                   3   \n",
       "1          141                 106           741                  23   \n",
       "2          134                  95           743                  34   \n",
       "3           89                  57           442                  15   \n",
       "4          124                  86           695                  28   \n",
       "5           68                  57           391                  22   \n",
       "6           30                  27           136                  11   \n",
       "7          146                  94           727                  24   \n",
       "8           35                  27           193                  14   \n",
       "9          142                 102           719                  28   \n",
       "\n",
       "   uppercase_words_count  titlecase_words_count  stopwords_count  \\\n",
       "0                      1                      6                7   \n",
       "1                      2                     13               69   \n",
       "2                      6                     15               59   \n",
       "3                      4                      9               45   \n",
       "4                      7                     15               60   \n",
       "5                      0                      3               18   \n",
       "6                      2                      4                8   \n",
       "7                     13                     27               77   \n",
       "8                      1                      3               11   \n",
       "9                     13                     16               66   \n",
       "\n",
       "   average_word_length  \n",
       "0             3.705882  \n",
       "1             4.248227  \n",
       "2             4.544776  \n",
       "3             3.977528  \n",
       "4             4.532258  \n",
       "5             4.735294  \n",
       "6             3.566667  \n",
       "7             3.979452  \n",
       "8             4.514286  \n",
       "9             4.070423  \n",
       "\n",
       "[10 rows x 24 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Label Encoding Drugname and Conditions\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "label_encoder_feat = {}\n",
    "for feature in ['drugName', 'condition']:\n",
    "    label_encoder_feat[feature] = LabelEncoder()\n",
    "    data[feature] = label_encoder_feat[feature].fit_transform(data[feature])\n",
    "\n",
    "data.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9e5134c7-48ca-4e63-b294-e328c0919f1c",
   "metadata": {},
   "source": [
    "### Define Features and Spliting Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "91a8f6f1-f9a8-4b82-b19c-b296ab22b5b6",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X = data[['condition', 'usefulCount', 'Day', 'Month', 'Year', 'sentiment', 'sentiment_clean_ss', \n",
    "                'words_count', 'unique_words_count', 'letters_count','punctuations_count', 'uppercase_words_count',\n",
    "                'titlecase_words_count', 'stopwords_count', 'average_word_length']]\n",
    "y = data['Review_Sentiment']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "3d2bcb12-e96a-4d8d-8aaf-5565b837d32f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(171095, 15) (42774, 15)\n",
      "(171095,) (42774,)\n"
     ]
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 42)\n",
    "print(X_train.shape, X_test.shape)\n",
    "print(y_train.shape, y_test.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f27d74d-5199-4df1-9ced-3d9215bcdf6f",
   "metadata": {},
   "source": [
    "### LGBM Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "eebda905-fe1b-4d6d-8215-234483939fb9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on Training Data:  0.9943189456150092\n",
      "Accuracy on Testing Data:  0.8986300088839014\n",
      "Precision on Testing Data:  0.9102068215963942\n",
      "Recall on Testing Data:  0.949101696610113\n",
      "F1 Score on Testing Data:  0.9292474381567783\n"
     ]
    }
   ],
   "source": [
    "from lightgbm import LGBMClassifier\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score\n",
    "\n",
    "clf = LGBMClassifier(\n",
    "        n_estimators = 20000,\n",
    "        learning_rate = 0.20,\n",
    "        num_leaves = 40,\n",
    "        subsample = .9,\n",
    "        max_depth = 14,\n",
    "        reg_alpha = .5,\n",
    "        reg_lambda = .5,\n",
    "        min_split_gain = .03,\n",
    "        min_child_weight = 3,\n",
    "        silent=-1,\n",
    "        verbose=-1,\n",
    "        )\n",
    "model_one = clf.fit(X_train, y_train)\n",
    "pred_one_test = model_one.predict(X_test)\n",
    "pred_one_train = model_one.predict(X_train)\n",
    "\n",
    "print (\"Accuracy on Training Data: \", accuracy_score(y_train, pred_one_train))\n",
    "print (\"Accuracy on Testing Data: \", accuracy_score(y_test, pred_one_test))\n",
    "print (\"Precision on Testing Data: \", precision_score(y_test, pred_one_test))\n",
    "print (\"Recall on Testing Data: \", recall_score(y_test, pred_one_test))\n",
    "print (\"F1 Score on Testing Data: \", f1_score(y_test, pred_one_test))\n",
    "\n",
    "\n",
    "lgbm_clf_train = accuracy_score(y_train, pred_one_train)\n",
    "lgbm_clf_test = accuracy_score(y_test, pred_one_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "358b1e1a-fb91-4140-9a7d-010366b331b3",
   "metadata": {},
   "source": [
    "### Confusion Matrix of LGBM Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "a662b2ca-f4a3-4230-9778-96ef9015b464",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cm = confusion_matrix(y_test, pred_one_test)\n",
    "\n",
    "sns.heatmap(cm, \n",
    "            annot=True,\n",
    "            fmt='g', \n",
    "            xticklabels=['Positive Sentiment','Negative Sentiment'],\n",
    "            yticklabels=['Positive Sentiment','Negative Sentiment'])\n",
    "plt.ylabel('Predicted',fontsize=13)\n",
    "plt.xlabel('Actual',fontsize=13)\n",
    "plt.title('Confusion Matrix of LGBM CLassifier',fontsize=17)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "83712a00-1786-421f-a83f-5f4ff8cabfa4",
   "metadata": {},
   "source": [
    "### XGBoost Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "4a40adaf-f90b-4837-a0b5-96944c1f9426",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on Training Data:  0.795137204477045\n",
      "Accuracy on Testing Data:  0.7776920559218217\n",
      "Precision on Testing Data:  0.7993630573248408\n",
      "Recall on Testing Data:  0.9119362687910403\n",
      "F1 Score on Testing Data:  0.851947000482663\n"
     ]
    }
   ],
   "source": [
    "from xgboost import XGBClassifier\n",
    "\n",
    "xgb = XGBClassifier()\n",
    "\n",
    "model_two = xgb.fit(X_train, y_train)\n",
    "pred_two_train = model_two.predict(X_train)\n",
    "pred_two_test = model_two.predict(X_test)\n",
    "\n",
    "print (\"Accuracy on Training Data: \", accuracy_score(y_train, pred_two_train))\n",
    "print (\"Accuracy on Testing Data: \", accuracy_score(y_test, pred_two_test))\n",
    "print (\"Precision on Testing Data: \", precision_score(y_test, pred_two_test))\n",
    "print (\"Recall on Testing Data: \", recall_score(y_test, pred_two_test))\n",
    "print (\"F1 Score on Testing Data: \", f1_score(y_test, pred_two_test))\n",
    "\n",
    "\n",
    "xgb_clf_train = accuracy_score(y_train, pred_two_train)\n",
    "xgb_clf_test = accuracy_score(y_test, pred_two_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4f8f0cbc-b0da-4594-a900-02ae88c7cdd0",
   "metadata": {},
   "source": [
    "### CatBoost Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "5ccc0c3b-d494-4521-be62-d9ae4f22ca04",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0:\tlearn: 0.5804071\ttotal: 214ms\tremaining: 21.2s\n",
      "1:\tlearn: 0.5413893\ttotal: 248ms\tremaining: 12.2s\n",
      "2:\tlearn: 0.5245131\ttotal: 279ms\tremaining: 9.03s\n",
      "3:\tlearn: 0.5169191\ttotal: 313ms\tremaining: 7.5s\n",
      "4:\tlearn: 0.5124387\ttotal: 344ms\tremaining: 6.53s\n",
      "5:\tlearn: 0.5090970\ttotal: 373ms\tremaining: 5.84s\n",
      "6:\tlearn: 0.5066599\ttotal: 408ms\tremaining: 5.43s\n",
      "7:\tlearn: 0.5047931\ttotal: 450ms\tremaining: 5.17s\n",
      "8:\tlearn: 0.5038500\ttotal: 480ms\tremaining: 4.85s\n",
      "9:\tlearn: 0.5028902\ttotal: 510ms\tremaining: 4.59s\n",
      "10:\tlearn: 0.5013339\ttotal: 538ms\tremaining: 4.35s\n",
      "11:\tlearn: 0.5004859\ttotal: 566ms\tremaining: 4.15s\n",
      "12:\tlearn: 0.4998158\ttotal: 594ms\tremaining: 3.97s\n",
      "13:\tlearn: 0.4984225\ttotal: 625ms\tremaining: 3.84s\n",
      "14:\tlearn: 0.4956493\ttotal: 660ms\tremaining: 3.74s\n",
      "15:\tlearn: 0.4948030\ttotal: 694ms\tremaining: 3.64s\n",
      "16:\tlearn: 0.4935550\ttotal: 731ms\tremaining: 3.57s\n",
      "17:\tlearn: 0.4929956\ttotal: 762ms\tremaining: 3.47s\n",
      "18:\tlearn: 0.4925840\ttotal: 788ms\tremaining: 3.36s\n",
      "19:\tlearn: 0.4919072\ttotal: 819ms\tremaining: 3.28s\n",
      "20:\tlearn: 0.4915338\ttotal: 851ms\tremaining: 3.2s\n",
      "21:\tlearn: 0.4909209\ttotal: 887ms\tremaining: 3.15s\n",
      "22:\tlearn: 0.4904599\ttotal: 920ms\tremaining: 3.08s\n",
      "23:\tlearn: 0.4897732\ttotal: 954ms\tremaining: 3.02s\n",
      "24:\tlearn: 0.4891829\ttotal: 983ms\tremaining: 2.95s\n",
      "25:\tlearn: 0.4887016\ttotal: 1.01s\tremaining: 2.88s\n",
      "26:\tlearn: 0.4874515\ttotal: 1.04s\tremaining: 2.82s\n",
      "27:\tlearn: 0.4872120\ttotal: 1.07s\tremaining: 2.75s\n",
      "28:\tlearn: 0.4867869\ttotal: 1.11s\tremaining: 2.71s\n",
      "29:\tlearn: 0.4862538\ttotal: 1.14s\tremaining: 2.67s\n",
      "30:\tlearn: 0.4857229\ttotal: 1.19s\tremaining: 2.66s\n",
      "31:\tlearn: 0.4850842\ttotal: 1.23s\tremaining: 2.6s\n",
      "32:\tlearn: 0.4846123\ttotal: 1.26s\tremaining: 2.56s\n",
      "33:\tlearn: 0.4842900\ttotal: 1.29s\tremaining: 2.5s\n",
      "34:\tlearn: 0.4835169\ttotal: 1.33s\tremaining: 2.47s\n",
      "35:\tlearn: 0.4827538\ttotal: 1.37s\tremaining: 2.44s\n",
      "36:\tlearn: 0.4823534\ttotal: 1.4s\tremaining: 2.38s\n",
      "37:\tlearn: 0.4818619\ttotal: 1.43s\tremaining: 2.33s\n",
      "38:\tlearn: 0.4813797\ttotal: 1.49s\tremaining: 2.32s\n",
      "39:\tlearn: 0.4810646\ttotal: 1.52s\tremaining: 2.27s\n",
      "40:\tlearn: 0.4807980\ttotal: 1.55s\tremaining: 2.23s\n",
      "41:\tlearn: 0.4804361\ttotal: 1.58s\tremaining: 2.19s\n",
      "42:\tlearn: 0.4800474\ttotal: 1.62s\tremaining: 2.15s\n",
      "43:\tlearn: 0.4796327\ttotal: 1.65s\tremaining: 2.1s\n",
      "44:\tlearn: 0.4793455\ttotal: 1.68s\tremaining: 2.05s\n",
      "45:\tlearn: 0.4788670\ttotal: 1.72s\tremaining: 2.01s\n",
      "46:\tlearn: 0.4783037\ttotal: 1.75s\tremaining: 1.97s\n",
      "47:\tlearn: 0.4779898\ttotal: 1.78s\tremaining: 1.93s\n",
      "48:\tlearn: 0.4778900\ttotal: 1.81s\tremaining: 1.88s\n",
      "49:\tlearn: 0.4775360\ttotal: 1.84s\tremaining: 1.84s\n",
      "50:\tlearn: 0.4773897\ttotal: 1.87s\tremaining: 1.8s\n",
      "51:\tlearn: 0.4768456\ttotal: 1.9s\tremaining: 1.75s\n",
      "52:\tlearn: 0.4767516\ttotal: 1.92s\tremaining: 1.71s\n",
      "53:\tlearn: 0.4764622\ttotal: 1.96s\tremaining: 1.67s\n",
      "54:\tlearn: 0.4761406\ttotal: 1.99s\tremaining: 1.63s\n",
      "55:\tlearn: 0.4758073\ttotal: 2.02s\tremaining: 1.59s\n",
      "56:\tlearn: 0.4755304\ttotal: 2.06s\tremaining: 1.55s\n",
      "57:\tlearn: 0.4751844\ttotal: 2.09s\tremaining: 1.51s\n",
      "58:\tlearn: 0.4749167\ttotal: 2.12s\tremaining: 1.47s\n",
      "59:\tlearn: 0.4744397\ttotal: 2.15s\tremaining: 1.43s\n",
      "60:\tlearn: 0.4740193\ttotal: 2.19s\tremaining: 1.4s\n",
      "61:\tlearn: 0.4738172\ttotal: 2.22s\tremaining: 1.36s\n",
      "62:\tlearn: 0.4729906\ttotal: 2.26s\tremaining: 1.33s\n",
      "63:\tlearn: 0.4726951\ttotal: 2.29s\tremaining: 1.29s\n",
      "64:\tlearn: 0.4721212\ttotal: 2.32s\tremaining: 1.25s\n",
      "65:\tlearn: 0.4717121\ttotal: 2.36s\tremaining: 1.21s\n",
      "66:\tlearn: 0.4713514\ttotal: 2.39s\tremaining: 1.18s\n",
      "67:\tlearn: 0.4710833\ttotal: 2.43s\tremaining: 1.14s\n",
      "68:\tlearn: 0.4707659\ttotal: 2.46s\tremaining: 1.1s\n",
      "69:\tlearn: 0.4704480\ttotal: 2.49s\tremaining: 1.07s\n",
      "70:\tlearn: 0.4699872\ttotal: 2.53s\tremaining: 1.03s\n",
      "71:\tlearn: 0.4696582\ttotal: 2.56s\tremaining: 994ms\n",
      "72:\tlearn: 0.4695787\ttotal: 2.58s\tremaining: 955ms\n",
      "73:\tlearn: 0.4692529\ttotal: 2.61s\tremaining: 918ms\n",
      "74:\tlearn: 0.4689140\ttotal: 2.65s\tremaining: 883ms\n",
      "75:\tlearn: 0.4685880\ttotal: 2.69s\tremaining: 848ms\n",
      "76:\tlearn: 0.4682471\ttotal: 2.72s\tremaining: 812ms\n",
      "77:\tlearn: 0.4681064\ttotal: 2.75s\tremaining: 775ms\n",
      "78:\tlearn: 0.4677155\ttotal: 2.78s\tremaining: 740ms\n",
      "79:\tlearn: 0.4674661\ttotal: 2.81s\tremaining: 703ms\n",
      "80:\tlearn: 0.4671636\ttotal: 2.84s\tremaining: 667ms\n",
      "81:\tlearn: 0.4668788\ttotal: 2.88s\tremaining: 633ms\n",
      "82:\tlearn: 0.4666738\ttotal: 2.91s\tremaining: 597ms\n",
      "83:\tlearn: 0.4665369\ttotal: 2.96s\tremaining: 564ms\n",
      "84:\tlearn: 0.4664006\ttotal: 2.98s\tremaining: 527ms\n",
      "85:\tlearn: 0.4661374\ttotal: 3.02s\tremaining: 491ms\n",
      "86:\tlearn: 0.4658592\ttotal: 3.04s\tremaining: 455ms\n",
      "87:\tlearn: 0.4656457\ttotal: 3.07s\tremaining: 419ms\n",
      "88:\tlearn: 0.4654083\ttotal: 3.11s\tremaining: 384ms\n",
      "89:\tlearn: 0.4650974\ttotal: 3.14s\tremaining: 349ms\n",
      "90:\tlearn: 0.4648973\ttotal: 3.17s\tremaining: 314ms\n",
      "91:\tlearn: 0.4646196\ttotal: 3.2s\tremaining: 278ms\n",
      "92:\tlearn: 0.4642901\ttotal: 3.24s\tremaining: 244ms\n",
      "93:\tlearn: 0.4639939\ttotal: 3.27s\tremaining: 209ms\n",
      "94:\tlearn: 0.4636382\ttotal: 3.31s\tremaining: 174ms\n",
      "95:\tlearn: 0.4634337\ttotal: 3.34s\tremaining: 139ms\n",
      "96:\tlearn: 0.4632419\ttotal: 3.37s\tremaining: 104ms\n",
      "97:\tlearn: 0.4629722\ttotal: 3.4s\tremaining: 69.5ms\n",
      "98:\tlearn: 0.4628878\ttotal: 3.43s\tremaining: 34.6ms\n",
      "99:\tlearn: 0.4625823\ttotal: 3.46s\tremaining: 0us\n",
      "Accuracy on Training Data:  0.7771822671615185\n",
      "Accuracy on Testing Data:  0.7707719642773647\n",
      "Precision on Testing Data:  0.7956262076233972\n",
      "Recall on Testing Data:  0.9058698043398553\n",
      "F1 Score on Testing Data:  0.8471765457691048\n"
     ]
    }
   ],
   "source": [
    "from catboost import CatBoostClassifier\n",
    "\n",
    "cat_clf = CatBoostClassifier(iterations = 100, learning_rate = 0.5)\n",
    "\n",
    "model_three = cat_clf.fit(X_train, y_train)\n",
    "pred_train_three = model_three.predict(X_train)\n",
    "pred_test_three = model_three.predict(X_test)\n",
    "\n",
    "print (\"Accuracy on Training Data: \", accuracy_score(y_train, pred_train_three))\n",
    "print (\"Accuracy on Testing Data: \", accuracy_score(y_test, pred_test_three))\n",
    "print (\"Precision on Testing Data: \", precision_score(y_test, pred_test_three))\n",
    "print (\"Recall on Testing Data: \", recall_score(y_test, pred_test_three))\n",
    "print (\"F1 Score on Testing Data: \", f1_score(y_test, pred_test_three))\n",
    "\n",
    "\n",
    "catb_clf_train = accuracy_score(y_train, pred_train_three)\n",
    "catb_clf_test = accuracy_score(y_test, pred_test_three)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "02adf8d1-3fc2-4564-8a9e-86175d659811",
   "metadata": {},
   "source": [
    "### Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "2de452ca-9366-4aec-bcca-c3bc6112cf1b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on Training Data:  0.9865630205441421\n",
      "Accuracy on Testing Data:  0.8767709356150932\n",
      "Precision on Testing Data:  0.897588424437299\n",
      "Recall on Testing Data:  0.9304689843671877\n",
      "F1 Score on Testing Data:  0.9137329994599106\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier  \n",
    "\n",
    "rf_clf = RandomForestClassifier(n_estimators= 5, criterion=\"entropy\")  \n",
    "\n",
    "model_four = rf_clf.fit(X_train, y_train)  \n",
    "pred_train_four = model_four.predict(X_train)\n",
    "pred_test_four = model_four.predict(X_test)\n",
    "\n",
    "print (\"Accuracy on Training Data: \", accuracy_score(y_train, pred_train_four))\n",
    "print (\"Accuracy on Testing Data: \", accuracy_score(y_test, pred_test_four))\n",
    "print (\"Precision on Testing Data: \", precision_score(y_test, pred_test_four))\n",
    "print (\"Recall on Testing Data: \", recall_score(y_test, pred_test_four))\n",
    "print (\"F1 Score on Testing Data: \", f1_score(y_test, pred_test_four))\n",
    "\n",
    "\n",
    "rf_train = accuracy_score(y_train, pred_train_four)\n",
    "rf_test = accuracy_score(y_test, pred_test_four)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03afb142-c60a-4baf-9aa2-8246dfee17e2",
   "metadata": {},
   "source": [
    "### Gradient Boosting Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "185262d5-b325-4633-b665-eccaf0222084",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on Training Data:  0.7596598381016394\n",
      "Accuracy on Testing Data:  0.76240239397765\n",
      "Precision on Testing Data:  0.7818698493890309\n",
      "Recall on Testing Data:  0.9171027632412253\n",
      "F1 Score on Testing Data:  0.8441042475188292\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import GradientBoostingClassifier\n",
    "\n",
    "gb_clf = GradientBoostingClassifier()\n",
    "\n",
    "model_five = gb_clf.fit(X_train, y_train)  \n",
    "pred_train_five = model_five.predict(X_train)\n",
    "pred_test_five = model_five.predict(X_test)\n",
    "\n",
    "print (\"Accuracy on Training Data: \", accuracy_score(y_train, pred_train_five))\n",
    "print (\"Accuracy on Testing Data: \", accuracy_score(y_test, pred_test_five))\n",
    "print (\"Precision on Testing Data: \", precision_score(y_test, pred_test_five))\n",
    "print (\"Recall on Testing Data: \", recall_score(y_test, pred_test_five))\n",
    "print (\"F1 Score on Testing Data: \", f1_score(y_test, pred_test_five))\n",
    "\n",
    "\n",
    "gb_clf_train = accuracy_score(y_train, pred_train_five)\n",
    "gb_clf_test = accuracy_score(y_test, pred_test_five)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "57f6be55-3d45-4fc8-8510-40fae87b20e7",
   "metadata": {},
   "source": [
    "### K-Nearest Neighbors"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "f5c3ab15-fc57-4a05-a0cf-87f1253ff770",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on Training Data:  0.7887664747654811\n",
      "Accuracy on Testing Data:  0.7071117968859587\n",
      "Precision on Testing Data:  0.7640622638658002\n",
      "Recall on Testing Data:  0.8426052464917836\n",
      "F1 Score on Testing Data:  0.8014139428716355\n"
     ]
    }
   ],
   "source": [
    "from sklearn.neighbors import KNeighborsClassifier \n",
    "\n",
    "knn = KNeighborsClassifier(n_neighbors = 7) \n",
    "\n",
    "model_six = knn.fit(X_train, y_train)  \n",
    "pred_train_six = model_six.predict(X_train)\n",
    "pred_test_six = model_six.predict(X_test)\n",
    "\n",
    "print (\"Accuracy on Training Data: \", accuracy_score(y_train, pred_train_six))\n",
    "print (\"Accuracy on Testing Data: \", accuracy_score(y_test, pred_test_six))\n",
    "print (\"Precision on Testing Data: \", precision_score(y_test, pred_test_six))\n",
    "print (\"Recall on Testing Data: \", recall_score(y_test, pred_test_six))\n",
    "print (\"F1 Score on Testing Data: \", f1_score(y_test, pred_test_six))\n",
    "\n",
    "\n",
    "knn_train = accuracy_score(y_train, pred_train_six)\n",
    "knn_test = accuracy_score(y_test, pred_test_six)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7199a4d3-181c-4e4e-ab1a-ef49e1e10913",
   "metadata": {},
   "source": [
    "### Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "8288e1da-973e-4a66-9294-0caee1cd8f57",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy on Training Data:  0.7046377743358953\n",
      "Accuracy on Testing Data:  0.7046804133352036\n",
      "Precision on Testing Data:  0.7091793723053207\n",
      "Recall on Testing Data:  0.981400619979334\n",
      "F1 Score on Testing Data:  0.8233731368326854\n"
     ]
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "\n",
    "lr = LogisticRegression()\n",
    "\n",
    "\n",
    "model_seven = lr.fit(X_train, y_train)  \n",
    "pred_train_seven = model_seven.predict(X_train)\n",
    "pred_test_seven = model_seven.predict(X_test)\n",
    "\n",
    "print (\"Accuracy on Training Data: \", accuracy_score(y_train, pred_train_seven))\n",
    "print (\"Accuracy on Testing Data: \", accuracy_score(y_test, pred_test_seven))\n",
    "print (\"Precision on Testing Data: \", precision_score(y_test, pred_test_seven))\n",
    "print (\"Recall on Testing Data: \", recall_score(y_test, pred_test_seven))\n",
    "print (\"F1 Score on Testing Data: \", f1_score(y_test, pred_test_seven))\n",
    "\n",
    "\n",
    "lr_train = accuracy_score(y_train, pred_train_seven)\n",
    "lr_test = accuracy_score(y_test, pred_test_seven)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "39f810b6-e431-4180-af3b-2c587ee25467",
   "metadata": {},
   "source": [
    "### Bar Plot of every applied models testing and training accuracy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "c4dac383-a30d-4601-944e-cd6633a1638f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "barWidth = 0.25\n",
    "fig = plt.subplots(figsize =(15, 8))\n",
    "\n",
    "# set height of bar\n",
    "train_accuracy = [lgbm_clf_train, xgb_clf_train, catb_clf_train, rf_train, gb_clf_train, knn_train, lr_train]\n",
    "test_accuracy = [lgbm_clf_test, xgb_clf_test, catb_clf_test, rf_test, gb_clf_test, knn_test, lr_test]\n",
    "\n",
    "# Set position of bar on X axis\n",
    "br1 = np.arange(len(test_accuracy))\n",
    "br2 = [x + barWidth for x in br1]\n",
    "\n",
    "# Make the plot\n",
    "plt.bar(br1, test_accuracy, color ='skyblue', width = barWidth, edgecolor ='black', label ='Testing Accuracy')\n",
    "plt.bar(br2, train_accuracy, color ='gray', width = barWidth, edgecolor ='black', label ='Training Accuracy')\n",
    "\n",
    "# Adding Xticks\n",
    "plt.xlabel('Machine Learning Models', fontsize = 15)\n",
    "plt.ylabel('Accuracy', fontsize = 15)\n",
    "plt.xticks([r + barWidth for r in range(len(test_accuracy))],\n",
    "\t\t['LGBM Clf', 'XGBoost Clf', 'CatBoost Clf', 'Random Forest Clf', 'Gradient Bossting Clf', 'K-Nearest Neighbors', 'Logistic Regression'])\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "259e0dfc-8eac-438f-9d17-251c4e5d43b4",
   "metadata": {},
   "source": [
    "### ROC Curve for all five models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "8b691d87-e09a-4ac1-a8d3-25df66c3cda6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve, roc_auc_score\n",
    "\n",
    "y_pred_prob1 = model_one.predict_proba(X_test)[:,1]\n",
    "fpr1 , tpr1, thresholds1 = roc_curve(y_test, y_pred_prob1)\n",
    "auc1 = round(roc_auc_score(y_test, y_pred_prob1), 4)\n",
    "\n",
    "\n",
    "y_pred_prob2 = model_two.predict_proba(X_test)[:,1]\n",
    "fpr2 , tpr2, thresholds2 = roc_curve(y_test, y_pred_prob2)\n",
    "auc2 = round(roc_auc_score(y_test, y_pred_prob2), 4)\n",
    "\n",
    "\n",
    "y_pred_prob3 = model_three.predict_proba(X_test)[:,1]\n",
    "fpr3 , tpr3, thresholds3 = roc_curve(y_test, y_pred_prob3)\n",
    "auc3 = round(roc_auc_score(y_test, y_pred_prob3), 4)\n",
    "\n",
    "\n",
    "y_pred_prob4 = model_four.predict_proba(X_test)[:,1]\n",
    "fpr4 , tpr4, thresholds4 = roc_curve(y_test, y_pred_prob4)\n",
    "auc4 = round(roc_auc_score(y_test, y_pred_prob4), 4)\n",
    "\n",
    "\n",
    "y_pred_prob5 = model_five.predict_proba(X_test)[:,1]\n",
    "fpr5 , tpr5, thresholds5 = roc_curve(y_test, y_pred_prob5)\n",
    "auc5 = round(roc_auc_score(y_test, y_pred_prob5), 4)\n",
    "\n",
    "y_pred_prob6 = model_six.predict_proba(X_test)[:,1]\n",
    "fpr6 , tpr6, thresholds6 = roc_curve(y_test, y_pred_prob6)\n",
    "auc6 = round(roc_auc_score(y_test, y_pred_prob6), 4)\n",
    "\n",
    "y_pred_prob7 = model_seven.predict_proba(X_test)[:,1]\n",
    "fpr7 , tpr7, thresholds7 = roc_curve(y_test, y_pred_prob7)\n",
    "auc7 = round(roc_auc_score(y_test, y_pred_prob7), 4)\n",
    "\n",
    "\n",
    "plt.plot(fpr1, tpr1, label= \"LGBM Clf, AUC = \" + str(auc1))\n",
    "plt.plot(fpr2, tpr2, label= \"XGBoost Clf, AUC = \" + str(auc2))\n",
    "plt.plot(fpr3, tpr3, label= \"CatBoost Clf, AUC = \" + str(auc3))\n",
    "plt.plot(fpr4, tpr4, label= \"Random Forest Clf, AUC = \" + str(auc4))\n",
    "plt.plot(fpr5, tpr5, label= \"Gradient Boosting Clf, AUC = \" + str(auc5))\n",
    "plt.plot(fpr6, tpr6, label= \"K-Nearest Neighbors, AUC = \" + str(auc6))\n",
    "plt.plot(fpr7, tpr7, label= \"Logistic Regression, AUC = \" + str(auc7))\n",
    "\n",
    "\n",
    "plt.legend()\n",
    "plt.xlabel(\"False Positive Rate (FPR)\")\n",
    "plt.ylabel(\"True Positive Rate (TPR)\")\n",
    "plt.title('Receiver Operating Characteristic Curve for five models')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ce44539-8d3b-416a-8041-95f9df49c2fd",
   "metadata": {},
   "source": [
    "### Prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "8da4541d-6434-4565-bdf1-fca7561c5490",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0.]\n",
      "Negative\n"
     ]
    }
   ],
   "source": [
    "# Making a predictive system\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "lb = LabelEncoder()\n",
    "\n",
    "# input_data = (314, 5, 20, 5, 2012, 0.111111, 0.111111, 17, 17, 79, 3, 1, 6,\t7, 3.705882) => Positive\n",
    "input_data = (314, 5, 7, 3, 2017, 0.111111, 0.111111, 30, 27, 136, 11, 2, 4, 8, 3.566667)\n",
    "# Changing the input_data to numpy array\n",
    "input_data_as_numpy_array = np.asarray(input_data)\n",
    "\n",
    "# Reshaping the array as we are predicting for one instance\n",
    "input_data_reshape = input_data_as_numpy_array.reshape(1,-1)\n",
    "\n",
    "prediction = clf.predict(input_data_reshape)\n",
    "print(prediction)\n",
    "\n",
    "if(prediction[0]==1):\n",
    "    print(\"Positive\")\n",
    "else:\n",
    "    print(\"Negative\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}