Final.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Reading the data and Converting to required Format"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "#importing the required packages\n",
    "import pandas as pd\n",
    "import numpy as np \n",
    "import seaborn as sns\n",
    "from wordcloud import WordCloud"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Reading in the data \n",
    "df=pd.read_csv('SMSSpamCollection.tsv',sep='\\t',header=None)\n",
    "df.columns=['Spam/Ham','EmailContent']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>Spam/Ham</th>\n",
       "      <th>EmailContent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ham</td>\n",
       "      <td>I've been searching for the right words to tha...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>spam</td>\n",
       "      <td>Free entry in 2 a wkly comp to win FA Cup fina...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ham</td>\n",
       "      <td>Nah I don't think he goes to usf, he lives aro...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ham</td>\n",
       "      <td>Even my brother is not like to speak with me. ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ham</td>\n",
       "      <td>I HAVE A DATE ON SUNDAY WITH WILL!!</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Spam/Ham                                       EmailContent\n",
       "0      ham  I've been searching for the right words to tha...\n",
       "1     spam  Free entry in 2 a wkly comp to win FA Cup fina...\n",
       "2      ham  Nah I don't think he goes to usf, he lives aro...\n",
       "3      ham  Even my brother is not like to speak with me. ...\n",
       "4      ham                I HAVE A DATE ON SUNDAY WITH WILL!!"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5568, 2)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "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>EmailContent</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Spam/Ham</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ham</th>\n",
       "      <td>4822</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>spam</th>\n",
       "      <td>746</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          EmailContent\n",
       "Spam/Ham              \n",
       "ham               4822\n",
       "spam               746"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Checking for number of spams\n",
    "groupedf=df.groupby('Spam/Ham').count()\n",
    "groupedf.head(15)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.Preprocessing the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Spam/Ham        0\n",
       "EmailContent    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Preprocessing by removing null values if present\n",
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Precprocessing by removing  duplicates\n",
    "cleaned_df=df.drop_duplicates(subset=['EmailContent'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5165\n",
      "(5165, 2)\n"
     ]
    }
   ],
   "source": [
    "print(len(cleaned_df))\n",
    "print(cleaned_df.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since dates are not mentioned,we cannot conclude whether its a duplicate entry.Spams generally have the same content but sent every day.\n",
    "Im not removing duplicates as of now but you can(its a personal choice)."
   ]
  },
  {
   "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>Spam/Ham</th>\n",
       "      <th>EmailContent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ham</td>\n",
       "      <td>I've been searching for the right words to tha...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>spam</td>\n",
       "      <td>Free entry in 2 a wkly comp to win FA Cup fina...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ham</td>\n",
       "      <td>Nah I don't think he goes to usf, he lives aro...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ham</td>\n",
       "      <td>Even my brother is not like to speak with me. ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Spam/Ham                                       EmailContent\n",
       "0      ham  I've been searching for the right words to tha...\n",
       "1     spam  Free entry in 2 a wkly comp to win FA Cup fina...\n",
       "2      ham  Nah I don't think he goes to usf, he lives aro...\n",
       "3      ham  Even my brother is not like to speak with me. ..."
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cleaned_df.head(4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.Tokenisation and Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~\n"
     ]
    }
   ],
   "source": [
    "#importing the required packages\n",
    "import string\n",
    "import re\n",
    "import nltk\n",
    "print(string.punctuation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package stopwords to\n",
      "[nltk_data]     C:\\Users\\b16z\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package stopwords is already up-to-date!\n",
      "[nltk_data] Downloading package wordnet to\n",
      "[nltk_data]     C:\\Users\\b16z\\AppData\\Roaming\\nltk_data...\n",
      "[nltk_data]   Package wordnet is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "nltk.download('stopwords')\n",
    "nltk.download('wordnet')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "stopwords=nltk.corpus.stopwords.words('english')\n",
    "ps=nltk.PorterStemmer()#Initialising Stemming\n",
    "wn = nltk.WordNetLemmatizer()#Initialising Lemmatisation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>Spam/Ham</th>\n",
       "      <th>EmailContent</th>\n",
       "      <th>body_text_filtered</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ham</td>\n",
       "      <td>I've been searching for the right words to tha...</td>\n",
       "      <td>[ive, searching, right, words, thank, breather...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>spam</td>\n",
       "      <td>Free entry in 2 a wkly comp to win FA Cup fina...</td>\n",
       "      <td>[free, entry, 2, wkly, comp, win, fa, cup, fin...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ham</td>\n",
       "      <td>Nah I don't think he goes to usf, he lives aro...</td>\n",
       "      <td>[nah, dont, think, goes, usf, lives, around, t...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ham</td>\n",
       "      <td>Even my brother is not like to speak with me. ...</td>\n",
       "      <td>[even, brother, like, speak, treat, like, aids...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ham</td>\n",
       "      <td>I HAVE A DATE ON SUNDAY WITH WILL!!</td>\n",
       "      <td>[date, sunday]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Spam/Ham                                       EmailContent  \\\n",
       "0      ham  I've been searching for the right words to tha...   \n",
       "1     spam  Free entry in 2 a wkly comp to win FA Cup fina...   \n",
       "2      ham  Nah I don't think he goes to usf, he lives aro...   \n",
       "3      ham  Even my brother is not like to speak with me. ...   \n",
       "4      ham                I HAVE A DATE ON SUNDAY WITH WILL!!   \n",
       "\n",
       "                                  body_text_filtered  \n",
       "0  [ive, searching, right, words, thank, breather...  \n",
       "1  [free, entry, 2, wkly, comp, win, fa, cup, fin...  \n",
       "2  [nah, dont, think, goes, usf, lives, around, t...  \n",
       "3  [even, brother, like, speak, treat, like, aids...  \n",
       "4                                     [date, sunday]  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def clean_text_trial(text):\n",
    "    text = \"\".join([word.lower() for word in text if word not in string.punctuation])#Removing punctuation\n",
    "    tokens = re.split('\\W+', text)#Using regex for tokenising-splitting by non word\n",
    "    text = [word for word in tokens if word not in stopwords]#Stemming and removing stopwords\n",
    "    return text\n",
    "\n",
    "df['body_text_filtered'] = df['EmailContent'].apply(lambda x: clean_text_trial(x.lower()))\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>Spam/Ham</th>\n",
       "      <th>EmailContent</th>\n",
       "      <th>body_text_filtered</th>\n",
       "      <th>body_text_lemmatized</th>\n",
       "      <th>body_text_stemmed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ham</td>\n",
       "      <td>I've been searching for the right words to tha...</td>\n",
       "      <td>[ive, searching, right, words, thank, breather...</td>\n",
       "      <td>[ive, searching, right, word, thank, breather,...</td>\n",
       "      <td>[ive, search, right, word, thank, breather, pr...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>spam</td>\n",
       "      <td>Free entry in 2 a wkly comp to win FA Cup fina...</td>\n",
       "      <td>[free, entry, 2, wkly, comp, win, fa, cup, fin...</td>\n",
       "      <td>[free, entry, 2, wkly, comp, win, fa, cup, fin...</td>\n",
       "      <td>[free, entri, 2, wkli, comp, win, fa, cup, fin...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ham</td>\n",
       "      <td>Nah I don't think he goes to usf, he lives aro...</td>\n",
       "      <td>[nah, dont, think, goes, usf, lives, around, t...</td>\n",
       "      <td>[nah, dont, think, go, usf, life, around, though]</td>\n",
       "      <td>[nah, dont, think, goe, usf, live, around, tho...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ham</td>\n",
       "      <td>Even my brother is not like to speak with me. ...</td>\n",
       "      <td>[even, brother, like, speak, treat, like, aids...</td>\n",
       "      <td>[even, brother, like, speak, treat, like, aid,...</td>\n",
       "      <td>[even, brother, like, speak, treat, like, aid,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ham</td>\n",
       "      <td>I HAVE A DATE ON SUNDAY WITH WILL!!</td>\n",
       "      <td>[date, sunday]</td>\n",
       "      <td>[date, sunday]</td>\n",
       "      <td>[date, sunday]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>ham</td>\n",
       "      <td>As per your request 'Melle Melle (Oru Minnamin...</td>\n",
       "      <td>[per, request, melle, melle, oru, minnaminungi...</td>\n",
       "      <td>[per, request, melle, melle, oru, minnaminungi...</td>\n",
       "      <td>[per, request, mell, mell, oru, minnaminungint...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>spam</td>\n",
       "      <td>WINNER!! As a valued network customer you have...</td>\n",
       "      <td>[winner, valued, network, customer, selected, ...</td>\n",
       "      <td>[winner, valued, network, customer, selected, ...</td>\n",
       "      <td>[winner, valu, network, custom, select, receiv...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>spam</td>\n",
       "      <td>Had your mobile 11 months or more? U R entitle...</td>\n",
       "      <td>[mobile, 11, months, u, r, entitled, update, l...</td>\n",
       "      <td>[mobile, 11, month, u, r, entitled, update, la...</td>\n",
       "      <td>[mobil, 11, month, u, r, entitl, updat, latest...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>ham</td>\n",
       "      <td>I'm gonna be home soon and i don't want to tal...</td>\n",
       "      <td>[im, gonna, home, soon, dont, want, talk, stuf...</td>\n",
       "      <td>[im, gonna, home, soon, dont, want, talk, stuf...</td>\n",
       "      <td>[im, gonna, home, soon, dont, want, talk, stuf...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>spam</td>\n",
       "      <td>SIX chances to win CASH! From 100 to 20,000 po...</td>\n",
       "      <td>[six, chances, win, cash, 100, 20000, pounds, ...</td>\n",
       "      <td>[six, chance, win, cash, 100, 20000, pound, tx...</td>\n",
       "      <td>[six, chanc, win, cash, 100, 20000, pound, txt...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Spam/Ham                                       EmailContent  \\\n",
       "0      ham  I've been searching for the right words to tha...   \n",
       "1     spam  Free entry in 2 a wkly comp to win FA Cup fina...   \n",
       "2      ham  Nah I don't think he goes to usf, he lives aro...   \n",
       "3      ham  Even my brother is not like to speak with me. ...   \n",
       "4      ham                I HAVE A DATE ON SUNDAY WITH WILL!!   \n",
       "5      ham  As per your request 'Melle Melle (Oru Minnamin...   \n",
       "6     spam  WINNER!! As a valued network customer you have...   \n",
       "7     spam  Had your mobile 11 months or more? U R entitle...   \n",
       "8      ham  I'm gonna be home soon and i don't want to tal...   \n",
       "9     spam  SIX chances to win CASH! From 100 to 20,000 po...   \n",
       "\n",
       "                                  body_text_filtered  \\\n",
       "0  [ive, searching, right, words, thank, breather...   \n",
       "1  [free, entry, 2, wkly, comp, win, fa, cup, fin...   \n",
       "2  [nah, dont, think, goes, usf, lives, around, t...   \n",
       "3  [even, brother, like, speak, treat, like, aids...   \n",
       "4                                     [date, sunday]   \n",
       "5  [per, request, melle, melle, oru, minnaminungi...   \n",
       "6  [winner, valued, network, customer, selected, ...   \n",
       "7  [mobile, 11, months, u, r, entitled, update, l...   \n",
       "8  [im, gonna, home, soon, dont, want, talk, stuf...   \n",
       "9  [six, chances, win, cash, 100, 20000, pounds, ...   \n",
       "\n",
       "                                body_text_lemmatized  \\\n",
       "0  [ive, searching, right, word, thank, breather,...   \n",
       "1  [free, entry, 2, wkly, comp, win, fa, cup, fin...   \n",
       "2  [nah, dont, think, go, usf, life, around, though]   \n",
       "3  [even, brother, like, speak, treat, like, aid,...   \n",
       "4                                     [date, sunday]   \n",
       "5  [per, request, melle, melle, oru, minnaminungi...   \n",
       "6  [winner, valued, network, customer, selected, ...   \n",
       "7  [mobile, 11, month, u, r, entitled, update, la...   \n",
       "8  [im, gonna, home, soon, dont, want, talk, stuf...   \n",
       "9  [six, chance, win, cash, 100, 20000, pound, tx...   \n",
       "\n",
       "                                   body_text_stemmed  \n",
       "0  [ive, search, right, word, thank, breather, pr...  \n",
       "1  [free, entri, 2, wkli, comp, win, fa, cup, fin...  \n",
       "2  [nah, dont, think, goe, usf, live, around, tho...  \n",
       "3  [even, brother, like, speak, treat, like, aid,...  \n",
       "4                                     [date, sunday]  \n",
       "5  [per, request, mell, mell, oru, minnaminungint...  \n",
       "6  [winner, valu, network, custom, select, receiv...  \n",
       "7  [mobil, 11, month, u, r, entitl, updat, latest...  \n",
       "8  [im, gonna, home, soon, dont, want, talk, stuf...  \n",
       "9  [six, chanc, win, cash, 100, 20000, pound, txt...  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Comparison between stemming and lemmatising\n",
    "def lemmatizing(tokenized_text):\n",
    "    text = [wn.lemmatize(word) for word in tokenized_text]\n",
    "    return text\n",
    "\n",
    "\n",
    "def stemming(tokenized_text):\n",
    "    text = [ps.stem(word) for word in tokenized_text]\n",
    "    return text\n",
    "\n",
    "df['body_text_lemmatized'] = df['body_text_filtered'].apply(lambda x: lemmatizing(x))\n",
    "\n",
    "df['body_text_stemmed'] = df['body_text_filtered'].apply(lambda x: stemming(x))\n",
    "\n",
    "df.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Generally stemming outperforms lemmatisation while lemmatisation becomes very effective in certain cases.In our scenario,stemming seems to give better results.If you observe 2nd row and 8th row ,the word entry,entitled,update is lemmatised as entri,entitl and updat.Stemming more or less better understanding our text."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "def clean_text(text):\n",
    "    text = \"\".join([word.lower() for word in text if word not in string.punctuation])#Removing punctuation\n",
    "    tokens = re.split('\\W+', text)#Using regex for tokenising-splitting by non word\n",
    "    text = [ps.stem(word) for word in tokens if word not in stopwords]#Stemming and removing stopwords\n",
    "    return text"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4.Feature Addition"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Adding in more features -actual content length removing white spaces and percent of punctuation in spam\n",
    "#This is based on the assumption that spams have higher body length and more number of punctuations\n",
    "def count_punct(text):\n",
    "    count = sum([1 for char in text if char in string.punctuation])\n",
    "    return round(count/(len(text) - text.count(\" \")), 3)*100\n",
    "\n",
    "df['body_len'] = df['EmailContent'].apply(lambda x: len(x) - x.count(\" \"))\n",
    "df['punct%'] = df['EmailContent'].apply(lambda x: count_punct(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>Spam/Ham</th>\n",
       "      <th>EmailContent</th>\n",
       "      <th>body_text_filtered</th>\n",
       "      <th>body_text_lemmatized</th>\n",
       "      <th>body_text_stemmed</th>\n",
       "      <th>body_len</th>\n",
       "      <th>punct%</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ham</td>\n",
       "      <td>I've been searching for the right words to tha...</td>\n",
       "      <td>[ive, searching, right, words, thank, breather...</td>\n",
       "      <td>[ive, searching, right, word, thank, breather,...</td>\n",
       "      <td>[ive, search, right, word, thank, breather, pr...</td>\n",
       "      <td>160</td>\n",
       "      <td>2.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>spam</td>\n",
       "      <td>Free entry in 2 a wkly comp to win FA Cup fina...</td>\n",
       "      <td>[free, entry, 2, wkly, comp, win, fa, cup, fin...</td>\n",
       "      <td>[free, entry, 2, wkly, comp, win, fa, cup, fin...</td>\n",
       "      <td>[free, entri, 2, wkli, comp, win, fa, cup, fin...</td>\n",
       "      <td>128</td>\n",
       "      <td>4.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ham</td>\n",
       "      <td>Nah I don't think he goes to usf, he lives aro...</td>\n",
       "      <td>[nah, dont, think, goes, usf, lives, around, t...</td>\n",
       "      <td>[nah, dont, think, go, usf, life, around, though]</td>\n",
       "      <td>[nah, dont, think, goe, usf, live, around, tho...</td>\n",
       "      <td>49</td>\n",
       "      <td>4.1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Spam/Ham                                       EmailContent  \\\n",
       "0      ham  I've been searching for the right words to tha...   \n",
       "1     spam  Free entry in 2 a wkly comp to win FA Cup fina...   \n",
       "2      ham  Nah I don't think he goes to usf, he lives aro...   \n",
       "\n",
       "                                  body_text_filtered  \\\n",
       "0  [ive, searching, right, words, thank, breather...   \n",
       "1  [free, entry, 2, wkly, comp, win, fa, cup, fin...   \n",
       "2  [nah, dont, think, goes, usf, lives, around, t...   \n",
       "\n",
       "                                body_text_lemmatized  \\\n",
       "0  [ive, searching, right, word, thank, breather,...   \n",
       "1  [free, entry, 2, wkly, comp, win, fa, cup, fin...   \n",
       "2  [nah, dont, think, go, usf, life, around, though]   \n",
       "\n",
       "                                   body_text_stemmed  body_len  punct%  \n",
       "0  [ive, search, right, word, thank, breather, pr...       160     2.5  \n",
       "1  [free, entri, 2, wkli, comp, win, fa, cup, fin...       128     4.7  \n",
       "2  [nah, dont, think, goe, usf, live, around, tho...        49     4.1  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lets evaluate if these features actually add some value to the model?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\b16z\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6521: MatplotlibDeprecationWarning: \n",
      "The 'normed' kwarg was deprecated in Matplotlib 2.1 and will be removed in 3.1. Use 'density' instead.\n",
      "  alternative=\"'density'\", removal=\"3.1\")\n",
      "C:\\Users\\b16z\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6521: MatplotlibDeprecationWarning: \n",
      "The 'normed' kwarg was deprecated in Matplotlib 2.1 and will be removed in 3.1. Use 'density' instead.\n",
      "  alternative=\"'density'\", removal=\"3.1\")\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "bins=np.linspace(0,200,40)\n",
    "plt.figure()\n",
    "plt.hist(df[df['Spam/Ham']=='ham']['body_len'],bins,label='Ham',normed=True)\n",
    "plt.hist(df[df['Spam/Ham']=='spam']['body_len'],bins,label='Spam',normed=True)\n",
    "plt.legend(loc='best')\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the histogram its pretty evident that Spam messages have higher length of text messages .And this feature has higher chances of improving our  models predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\b16z\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6521: MatplotlibDeprecationWarning: \n",
      "The 'normed' kwarg was deprecated in Matplotlib 2.1 and will be removed in 3.1. Use 'density' instead.\n",
      "  alternative=\"'density'\", removal=\"3.1\")\n",
      "C:\\Users\\b16z\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6521: MatplotlibDeprecationWarning: \n",
      "The 'normed' kwarg was deprecated in Matplotlib 2.1 and will be removed in 3.1. Use 'density' instead.\n",
      "  alternative=\"'density'\", removal=\"3.1\")\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "bins=np.linspace(0,200,40)\n",
    "plt.figure()\n",
    "plt.hist(df[df['Spam/Ham']=='ham']['punct%'],bins,alpha=.5,label='Ham',normed=True)\n",
    "plt.hist(df[df['Spam/Ham']=='spam']['punct%'],bins,alpha=.5,label='Spam',normed=True)\n",
    "plt.legend(loc='best')\n",
    "plt.show();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the histogram its observed that though there is slight difference between punctuation usage in ham and spam ,its not very distinct.Lets include the feature as of now because the number of features are not very high in this specific dataset. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>Spam/Ham</th>\n",
       "      <th>EmailContent</th>\n",
       "      <th>body_text_filtered</th>\n",
       "      <th>body_text_lemmatized</th>\n",
       "      <th>body_text_stemmed</th>\n",
       "      <th>body_len</th>\n",
       "      <th>punct%</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ham</td>\n",
       "      <td>I've been searching for the right words to tha...</td>\n",
       "      <td>[ive, searching, right, words, thank, breather...</td>\n",
       "      <td>[ive, searching, right, word, thank, breather,...</td>\n",
       "      <td>[ive, search, right, word, thank, breather, pr...</td>\n",
       "      <td>160</td>\n",
       "      <td>2.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>spam</td>\n",
       "      <td>Free entry in 2 a wkly comp to win FA Cup fina...</td>\n",
       "      <td>[free, entry, 2, wkly, comp, win, fa, cup, fin...</td>\n",
       "      <td>[free, entry, 2, wkly, comp, win, fa, cup, fin...</td>\n",
       "      <td>[free, entri, 2, wkli, comp, win, fa, cup, fin...</td>\n",
       "      <td>128</td>\n",
       "      <td>4.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ham</td>\n",
       "      <td>Nah I don't think he goes to usf, he lives aro...</td>\n",
       "      <td>[nah, dont, think, goes, usf, lives, around, t...</td>\n",
       "      <td>[nah, dont, think, go, usf, life, around, though]</td>\n",
       "      <td>[nah, dont, think, goe, usf, live, around, tho...</td>\n",
       "      <td>49</td>\n",
       "      <td>4.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ham</td>\n",
       "      <td>Even my brother is not like to speak with me. ...</td>\n",
       "      <td>[even, brother, like, speak, treat, like, aids...</td>\n",
       "      <td>[even, brother, like, speak, treat, like, aid,...</td>\n",
       "      <td>[even, brother, like, speak, treat, like, aid,...</td>\n",
       "      <td>62</td>\n",
       "      <td>3.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ham</td>\n",
       "      <td>I HAVE A DATE ON SUNDAY WITH WILL!!</td>\n",
       "      <td>[date, sunday]</td>\n",
       "      <td>[date, sunday]</td>\n",
       "      <td>[date, sunday]</td>\n",
       "      <td>28</td>\n",
       "      <td>7.1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Spam/Ham                                       EmailContent  \\\n",
       "0      ham  I've been searching for the right words to tha...   \n",
       "1     spam  Free entry in 2 a wkly comp to win FA Cup fina...   \n",
       "2      ham  Nah I don't think he goes to usf, he lives aro...   \n",
       "3      ham  Even my brother is not like to speak with me. ...   \n",
       "4      ham                I HAVE A DATE ON SUNDAY WITH WILL!!   \n",
       "\n",
       "                                  body_text_filtered  \\\n",
       "0  [ive, searching, right, words, thank, breather...   \n",
       "1  [free, entry, 2, wkly, comp, win, fa, cup, fin...   \n",
       "2  [nah, dont, think, goes, usf, lives, around, t...   \n",
       "3  [even, brother, like, speak, treat, like, aids...   \n",
       "4                                     [date, sunday]   \n",
       "\n",
       "                                body_text_lemmatized  \\\n",
       "0  [ive, searching, right, word, thank, breather,...   \n",
       "1  [free, entry, 2, wkly, comp, win, fa, cup, fin...   \n",
       "2  [nah, dont, think, go, usf, life, around, though]   \n",
       "3  [even, brother, like, speak, treat, like, aid,...   \n",
       "4                                     [date, sunday]   \n",
       "\n",
       "                                   body_text_stemmed  body_len  punct%  \n",
       "0  [ive, search, right, word, thank, breather, pr...       160     2.5  \n",
       "1  [free, entri, 2, wkli, comp, win, fa, cup, fin...       128     4.7  \n",
       "2  [nah, dont, think, goe, usf, live, around, tho...        49     4.1  \n",
       "3  [even, brother, like, speak, treat, like, aid,...        62     3.2  \n",
       "4                                     [date, sunday]        28     7.1  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "spam_words = df[df['Spam/Ham'] == 'spam']['body_text_stemmed']\n",
    "ham_words = df[df['Spam/Ham'] == 'ham']['body_text_stemmed']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "comment_words = ' '\n",
    "for val in spam_words: \n",
    "    for words in val: \n",
    "        comment_words = comment_words + words + ' '\n",
    "\n",
    "wordcloud = WordCloud(width = 800, height = 800, \n",
    "                background_color ='white', \n",
    "                stopwords = stopwords, \n",
    "                min_font_size = 10).generate(comment_words)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (10, 8), facecolor = 'k')\n",
    "plt.imshow(wordcloud)\n",
    "plt.axis('off')\n",
    "plt.tight_layout(pad = 0)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "comment_words = ' '\n",
    "for val in ham_words: \n",
    "    for words in val: \n",
    "        comment_words = comment_words + words + ' '\n",
    "\n",
    "wordcloud = WordCloud(width = 800, height = 800, \n",
    "                background_color ='white', \n",
    "                stopwords = stopwords, \n",
    "                min_font_size = 10).generate(comment_words)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (10, 8), facecolor = 'k')\n",
    "plt.imshow(wordcloud)\n",
    "plt.axis('off')\n",
    "plt.tight_layout(pad = 0)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the pictorial representation,its pretty clear that there is clear distinction between words in a spam mail which makes our task easier."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5.Train-Test Split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(df[['EmailContent','body_len','punct%']], df['Spam/Ham'], test_size=0.3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6.Vectorisation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "TfidfVectoriser is a modification of Count Vectoriser where it considers length of document(in our case text message) and occurence of words.N-grams are generally for obtaining group of words applied within these two vectorisation methods as an additional criteria.Here I am considering tfidf vectorisation technique because I have also added an additional feature body_len based on the hypothesis that spam messages have more text content,so length of document is important feature and taken into consideration while vectorising by tfidf method.\n",
    "\n",
    "From model accuracy perspective,tfidf and count vectoriser techniques didnt not much have impact because we have already added body_len as an additional feature while modelling. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "tfidf_vect = TfidfVectorizer(analyzer=clean_text)#Initialising the model\n",
    "tfidf_vect_fit = tfidf_vect.fit(X_train['EmailContent'])#Fitting the train set to model\n",
    "\n",
    "tfidf_train = tfidf_vect_fit.transform(X_train['EmailContent'])#Transforming the train set\n",
    "tfidf_test = tfidf_vect_fit.transform(X_test['EmailContent'])#Transforming the test set\n",
    "\n",
    "\n",
    "#Concatenating the transformed test and train sets with additional features constructed earlier\n",
    "X_train_vect = pd.concat([X_train[['body_len', 'punct%']].reset_index(drop=True), \n",
    "           pd.DataFrame(tfidf_train.toarray())], axis=1)\n",
    "X_test_vect = pd.concat([X_test[['body_len', 'punct%']].reset_index(drop=True), \n",
    "           pd.DataFrame(tfidf_test.toarray())], axis=1)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>body_len</th>\n",
       "      <th>punct%</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>...</th>\n",
       "      <th>6674</th>\n",
       "      <th>6675</th>\n",
       "      <th>6676</th>\n",
       "      <th>6677</th>\n",
       "      <th>6678</th>\n",
       "      <th>6679</th>\n",
       "      <th>6680</th>\n",
       "      <th>6681</th>\n",
       "      <th>6682</th>\n",
       "      <th>6683</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>190</td>\n",
       "      <td>8.4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>59</td>\n",
       "      <td>1.7</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>79</td>\n",
       "      <td>2.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>42</td>\n",
       "      <td>2.4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>39</td>\n",
       "      <td>20.5</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 6686 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   body_len  punct%    0    1    2    3    4    5    6    7  ...  6674  6675  \\\n",
       "0       190     8.4  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...   0.0   0.0   \n",
       "1        59     1.7  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...   0.0   0.0   \n",
       "2        79     2.5  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...   0.0   0.0   \n",
       "3        42     2.4  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...   0.0   0.0   \n",
       "4        39    20.5  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...   0.0   0.0   \n",
       "\n",
       "   6676  6677  6678  6679  6680  6681  6682  6683  \n",
       "0   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0  \n",
       "1   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0  \n",
       "2   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0  \n",
       "3   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0  \n",
       "4   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0  \n",
       "\n",
       "[5 rows x 6686 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_vect.head()"
   ]
  },
  {
   "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>body_len</th>\n",
       "      <th>punct%</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>...</th>\n",
       "      <th>6674</th>\n",
       "      <th>6675</th>\n",
       "      <th>6676</th>\n",
       "      <th>6677</th>\n",
       "      <th>6678</th>\n",
       "      <th>6679</th>\n",
       "      <th>6680</th>\n",
       "      <th>6681</th>\n",
       "      <th>6682</th>\n",
       "      <th>6683</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>119</td>\n",
       "      <td>3.4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.252762</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>130</td>\n",
       "      <td>3.8</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>74</td>\n",
       "      <td>5.4</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>28</td>\n",
       "      <td>3.6</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>49</td>\n",
       "      <td>6.1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 6686 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   body_len  punct%    0    1    2    3    4    5    6    7  ...      6674  \\\n",
       "0       119     3.4  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.252762   \n",
       "1       130     3.8  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.000000   \n",
       "2        74     5.4  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.000000   \n",
       "3        28     3.6  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.000000   \n",
       "4        49     6.1  0.0  0.0  0.0  0.0  0.0  0.0  0.0  0.0  ...  0.000000   \n",
       "\n",
       "   6675  6676  6677  6678  6679  6680  6681  6682  6683  \n",
       "0   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0  \n",
       "1   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0  \n",
       "2   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0  \n",
       "3   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0  \n",
       "4   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0   0.0  \n",
       "\n",
       "[5 rows x 6686 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test_vect.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 7.Training Different models and Evaluation of Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier\n",
    "from sklearn.metrics import precision_recall_fscore_support as score\n",
    "import time\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.metrics import classification_report\n",
    "import seaborn as sns\n",
    "from sklearn.ensemble import AdaBoostClassifier"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 7.1 KNN Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "i=np.arange(1,20)\n",
    "accuracy_test=[]\n",
    "accuracy_train=[]\n",
    "for x in i:\n",
    "    KNN_model = KNeighborsClassifier(n_neighbors=x,n_jobs=-1)#Instantiate\n",
    "    KNN_model.fit(X_train_vect, y_train)#Fit\n",
    "    prediction1 = KNN_model.predict(X_test_vect)#Predict on test set\n",
    "    acc_test = accuracy_score(y_test, prediction1)\n",
    "    accuracy_test.append(acc_test)\n",
    "    prediction2 = KNN_model.predict(X_train_vect)#Predict on train set\n",
    "    acc_train = accuracy_score(y_train,prediction2)\n",
    "    accuracy_train.append(acc_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plotting the accuracy versus number of neighbors value to find best value of n_neighbors\n",
    "plt.figure()\n",
    "plt.plot(accuracy_train,label='train')\n",
    "plt.plot(accuracy_test,label='test')\n",
    "plt.xlabel('Number of Neighbors')\n",
    "plt.ylabel('Accuracy Score')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train:0.9160892994611239\n",
      "Test:0.9054458408138839\n"
     ]
    }
   ],
   "source": [
    "#After finding ideal N-neighbors,lets evaluate the model.\n",
    "\n",
    "KNN_model = KNeighborsClassifier(n_neighbors=10,n_jobs=-1)\n",
    "KNN_model.fit(X_train_vect, y_train)\n",
    "y_pred=KNN_model.predict(X_train_vect)\n",
    "print(f\"Train:{accuracy_score(y_train, y_pred)}\")\n",
    "y_pred=KNN_model.predict(X_test_vect)\n",
    "print(f\"Test:{accuracy_score(y_test, y_pred)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 7.2 Decision Tree Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "#i is the array storing maximum depth values\n",
    "i=range(25,45)\n",
    "accuracy_test=[]\n",
    "accuracy_train=[]\n",
    "for x in i:\n",
    "    #Instantiate the model\n",
    "    DT_model = DecisionTreeClassifier(max_depth=x)\n",
    "    #Fit the model \n",
    "    DT_model.fit(X_train_vect, y_train)\n",
    "    #Evaluate the model on test dataset\n",
    "    acc_test=DT_model.score(X_test_vect,y_test)\n",
    "    accuracy_test.append(acc_test)\n",
    "    #Evaluate the model on train dataset\n",
    "    acc_train=DT_model.score(X_train_vect, y_train)\n",
    "    accuracy_train.append(acc_train)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Lets visualise the accuarcy versus depth of DT tree to estimate ideal value\n",
    "plt.figure()\n",
    "plt.plot(accuracy_train,label='train')\n",
    "plt.plot(accuracy_test,label='test')\n",
    "plt.xlabel('Number of Neighbors')\n",
    "plt.ylabel('Accuracy Score')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train:0.9774185270721067\n",
      "Test:0.9533213644524237\n"
     ]
    }
   ],
   "source": [
    "#Instantiate the model\n",
    "DT_model = DecisionTreeClassifier(max_depth=10)\n",
    "#Fit the model\n",
    "DT_model.fit(X_train_vect, y_train)\n",
    "#Predict on train and test set\n",
    "y_pred_DT_train=DT_model.predict(X_train_vect)\n",
    "y_pred_DT_test=DT_model.predict(X_test_vect)\n",
    "#Evaluate the model\n",
    "print(f\"Train:{accuracy_score(y_train, y_pred_DT_train)}\")\n",
    "print(f\"Test:{accuracy_score(y_test, y_pred_DT_test)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 7.3 SVM-Linear and RBF Kernel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\b16z\\Anaconda3\\lib\\site-packages\\sklearn\\svm\\base.py:196: FutureWarning: The default value of gamma will change from 'auto' to 'scale' in version 0.22 to account better for unscaled features. Set gamma explicitly to 'auto' or 'scale' to avoid this warning.\n",
      "  \"avoid this warning.\", FutureWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVM-Rbf\n",
      "Train:0.8917115730048756\n",
      "Test:0.895870736086176\n",
      "\n",
      "SVM-Linear\n",
      "Train:0.9961508852963818\n",
      "Test:0.9814482345900658\n"
     ]
    }
   ],
   "source": [
    "from sklearn.svm import SVC\n",
    "\n",
    "#Instantiate the models\n",
    "my_kernel_SVM = SVC()\n",
    "my_linear_kernel_SVM = SVC(kernel='linear')\n",
    "\n",
    "\n",
    "#Fit the data\n",
    "my_linear_kernel_SVM.fit(X_train_vect,y_train)\n",
    "my_kernel_SVM.fit(X_train_vect,y_train)\n",
    "\n",
    "#Evaluate/Score\n",
    "print('SVM-Rbf')\n",
    "print(f\"Train:{my_kernel_SVM.score(X_train_vect,y_train)}\")\n",
    "print(f\"Test:{my_kernel_SVM.score(X_test_vect,y_test)}\")\n",
    "print('\\nSVM-Linear')\n",
    "print(f\"Train:{my_linear_kernel_SVM.score(X_train_vect,y_train)}\")\n",
    "print(f\"Test:{my_linear_kernel_SVM.score(X_test_vect,y_test)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x17ad4a4dfd0>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "y_pred=my_linear_kernel_SVM.predict(X_test_vect)\n",
    "CT_SVM=confusion_matrix(y_test,y_pred)\n",
    "sns.heatmap(CT_SVM,annot=True,cmap='YlGnBu')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since here more than accuracy,precison is what matters(we do not want actual emails to fall in spam category),we evaluate models with decent accuracy based on precision further."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9692307692307692"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import precision_score\n",
    "precision_score(y_test,y_pred,pos_label='spam')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 7.4 Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "i=range(100,200)\n",
    "accuracy_test=[]\n",
    "accuracy_train=[]\n",
    "for x in i:\n",
    "    #Instantiate the model\n",
    "    RF_model = RandomForestClassifier(n_estimators=x)\n",
    "    #Fit the model \n",
    "    RF_model.fit(X_train_vect, y_train)\n",
    "    #Evaluate the model on test dataset\n",
    "    acc_test=RF_model.score(X_test_vect,y_test)\n",
    "    accuracy_test.append(acc_test)\n",
    "    #Evaluate the model on train dataset\n",
    "    acc_train=RF_model.score(X_train_vect, y_train)\n",
    "    accuracy_train.append(acc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Lets visualise the accuarcy versus depth of DT tree to estimate ideal value\n",
    "plt.figure()\n",
    "plt.plot(accuracy_train,label='train')\n",
    "plt.plot(accuracy_test,label='test')\n",
    "plt.xlabel('Number of Neighbors')\n",
    "plt.ylabel('Accuracy Score')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train:1.0\n",
      "Test:0.9796529024536206\n"
     ]
    }
   ],
   "source": [
    "rf = RandomForestClassifier(n_estimators=150, max_depth=None, n_jobs=-1)\n",
    "rf_model = rf.fit(X_train_vect, y_train)\n",
    "y_pred = rf_model.predict(X_test_vect)\n",
    "\n",
    "\n",
    "print(f\"Train:{rf_model.score(X_train_vect,y_train)}\")\n",
    "print(f\"Test:{rf_model.score(X_test_vect,y_test)}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x17ad55bc6a0>"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "y_pred=rf_model.predict(X_test_vect)\n",
    "CT_RF=confusion_matrix(y_test,y_pred)\n",
    "sns.heatmap(CT_RF,annot=True,cmap='YlGnBu')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9891304347826086"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import precision_score\n",
    "precision_score(y_test,y_pred,pos_label='spam')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 7.5 Gradient Boosting "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "i=range(100,200)\n",
    "accuracy_test=[]\n",
    "accuracy_train=[]\n",
    "for x in i:\n",
    "    #Instantiate the model\n",
    "    GB_model = GradientBoostingClassifier(n_estimators=x)\n",
    "    #Fit the model \n",
    "    GB_model.fit(X_train_vect, y_train)\n",
    "    #Evaluate the model on test dataset\n",
    "    acc_test=GB_model.score(X_test_vect,y_test)\n",
    "    accuracy_test.append(acc_test)\n",
    "    #Evaluate the model on train dataset\n",
    "    acc_train=GB_model.score(X_train_vect, y_train)\n",
    "    accuracy_train.append(acc_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Lets visualise the accuarcy versus depth of DT tree to estimate ideal value\n",
    "plt.figure()\n",
    "plt.plot(accuracy_train,label='train')\n",
    "plt.plot(accuracy_test,label='test')\n",
    "plt.xlabel('Number of Neighbors')\n",
    "plt.ylabel('Accuracy Score')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train:1.0\n",
      "Test:0.9706762417713943\n"
     ]
    }
   ],
   "source": [
    "gb = GradientBoostingClassifier(n_estimators=50, max_depth=11)\n",
    "gb_model = gb.fit(X_train_vect, y_train)\n",
    "y_pred = gb_model.predict(X_test_vect)\n",
    "\n",
    "\n",
    "print(f\"Train:{gb_model.score(X_train_vect,y_train)}\")\n",
    "print(f\"Test:{gb_model.score(X_test_vect,y_test)}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.914572864321608"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import precision_score\n",
    "precision_score(y_test,y_pred,pos_label='spam')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 7.6 AdaBoost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_scores = []\n",
    "test_scores = []\n",
    "for i in np.arange(1,100):\n",
    "    my_adaboost = AdaBoostClassifier(n_estimators=i)\n",
    "    my_adaboost.fit(X_train_vect,y_train)\n",
    "    train_scores.append(my_adaboost.score(X_train_vect,y_train))\n",
    "    test_scores.append(my_adaboost.score(X_test_vect,y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "plt.plot(np.arange(1,100),train_scores,label = 'train')\n",
    "plt.plot(np.arange(1,100),test_scores,label = 'test')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train:0.9989735694123685\n",
      "Test:0.9796529024536206\n"
     ]
    }
   ],
   "source": [
    "my_adaboost = AdaBoostClassifier(n_estimators=150)\n",
    "my_adaboost.fit(X_train_vect,y_train)\n",
    "\n",
    "print(f\"Train:{my_adaboost.score(X_train_vect,y_train)}\")\n",
    "print(f\"Test:{my_adaboost.score(X_test_vect,y_test)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.914572864321608"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import precision_score\n",
    "precision_score(y_test,y_pred,pos_label='spam')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 7.7 XGBoost "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "from xgboost import XGBClassifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_scores = []\n",
    "test_scores = []\n",
    "for i in np.arange(1,100,4):\n",
    "    XGB_model = XGBClassifier(n_estimators=i)\n",
    "    XGB_model.fit(X_train_vect,y_train)\n",
    "    train_scores.append(XGB_model.score(X_train_vect,y_train))\n",
    "    test_scores.append(XGB_model.score(X_test_vect,y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "plt.plot(np.arange(1,100,4),train_scores,label = 'train')\n",
    "plt.plot(np.arange(1,100,4),test_scores,label = 'test')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train:0.9751090582499359\n",
      "Test:0.9694793536804309\n"
     ]
    }
   ],
   "source": [
    "XGB_model = XGBClassifier(n_estimators=100)\n",
    "XGB_model.fit(X_train_vect,y_train)\n",
    "\n",
    "print(f\"Train:{XGB_model.score(X_train_vect,y_train)}\")\n",
    "print(f\"Test:{XGB_model.score(X_test_vect,y_test)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.914572864321608"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import precision_score\n",
    "precision_score(y_test,y_pred,pos_label='spam')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 7.8 Naive Bayes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train\n",
      "0.953810623556582\n",
      "\n",
      "Test\n",
      "0.8886894075403949\n"
     ]
    }
   ],
   "source": [
    "from sklearn.naive_bayes import MultinomialNB, GaussianNB, BernoulliNB\n",
    "\n",
    "Gaus_model=GaussianNB()\n",
    "Gaus_model.fit(X_train_vect,y_train)\n",
    "print(\"Train\")\n",
    "print(Gaus_model.score(X_train_vect,y_train))\n",
    "print(\"\\nTest\")\n",
    "print(Gaus_model.score(X_test_vect,y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train\n",
      "0.9856299717731588\n",
      "\n",
      "Test\n",
      "0.9670855774985039\n"
     ]
    }
   ],
   "source": [
    "Berno_model=BernoulliNB()\n",
    "Berno_model.fit(X_train_vect,y_train)\n",
    "print(\"Train\")\n",
    "print(Berno_model.score(X_train_vect,y_train))\n",
    "print(\"\\nTest\")\n",
    "print(Berno_model.score(X_test_vect,y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train\n",
      "0.9379009494482936\n",
      "\n",
      "Test\n",
      "0.9275882704967086\n"
     ]
    }
   ],
   "source": [
    "Multi=MultinomialNB()\n",
    "Multi.fit(X_train_vect,y_train)\n",
    "print(\"Train\")\n",
    "print(Multi.score(X_train_vect,y_train))\n",
    "print(\"\\nTest\")\n",
    "print(Multi.score(X_test_vect,y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since we know approximately which models are better suited for our dataset,you can dig deeper and fine tune its hyperparamaters.From the above models,lets build an ensemble using highest accurate models to check if it can improve the accuracy further.Using accuracy metric,the models which perform well are:\n",
    "* Random Forest\n",
    "* Linear SVM\n",
    "* XG Boost\n",
    "* Gradient Boost\n",
    "* AdaBoost\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "But from precision perspective we can conclude that Random Forest and Linear SVM perform well."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 7.9 Ensemble "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From models evaluated above,lets form an ensemble of best performing models."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9994867847061842\n",
      "0.9838420107719928\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import VotingClassifier\n",
    "SVM_model = SVC(kernel = 'linear',probability=True)\n",
    "RF_model= RandomForestClassifier(n_estimators=150, max_depth=None, n_jobs=-1)\n",
    "\n",
    "\n",
    "#Build the ensemble\n",
    "ensemble = VotingClassifier(estimators=[('RF Model', RF_model),\\\n",
    "                                        ('SVM Model Linear', SVM_model)],\\\n",
    "                                        voting=\"soft\",\\\n",
    "                                        weights = [2,1])\n",
    "\n",
    "ensemble.fit(X_train_vect,y_train)\n",
    "print(ensemble.score(X_train_vect,y_train))\n",
    "print(ensemble.score(X_test_vect,y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You can again play with weights of models in  an ensemble model or try grid search for choosing best model and its hyper parameters.I would restrict myself with these models.But feel free to try more models and techniques."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x17a90cb2710>"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWcAAAD8CAYAAACrbmW5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAFsxJREFUeJzt3Xl4VdW9xvHv74TBARki86DFSFDRCyrFqVp9rAhUJisKrYJDbxSlitZbQURFoc6IqNUbLYqtFQG1YkEseB1a68BwqYKgRkQIICjzTBLW/eNsuEcSkpPkJGdl+X6eZz85WWcPayvPy+K3197bnHOIiIhfYunugIiIFKdwFhHxkMJZRMRDCmcREQ8pnEVEPKRwFhHxkMJZRMRDCmcREQ8pnEVEPFSrqg9w8BEDdAuiFLNj+ah0d0G8lG2V3UN5MmfH8hcqfbyqopGziIiHqnzkLCJSnczCGHMqnEUkKDELI9bCOAsRkYhGziIiHjLz9hpfuSicRSQwGjmLiHhHZQ0REQ8pnEVEPKTZGiIiHtLIWUTEQwpnEREPGZpKJyLiHY2cRUQ8FIuFEWthnIWIyD4aOYuIeEdlDRERDymcRUQ8ZCpriIj4RyNnEREPxWIZ6e5CSiicRSQoKmuIiHhIZQ0REQ8pnEVEPKSyhoiIh0y3b4uI+CeUF7yGMf4XEYkYsaSXMvdlNsHM1prZwoS2B8xsiZl9bGavmFnDqP1HZrbDzBZEy5MJ25xsZp+YWZ6Zjbck/gZROItIUMxiSS9JeBbotl/bLOB459x/AJ8DwxO++9I51ylarklofwLIAdpFy/77LEbhLCJhMUt+KYNz7l1g/X5tf3fOFUa/fgC0Lr071gKo75x73znngOeAPmUdW+EsImGJlWOpvCuB1xN+b2tm/2tm75jZmVFbKyA/YZ38qK1UuiAoImGJJZ+6ZpZDvNywV65zLjfJbUcAhcDzUdNq4Ajn3DozOxn4q5l1gBLfm+XK2r/CWUTCUo4RcRTESYVxIjMbBFwAnBuVKnDO7QJ2RZ/nmdmXQDbxkXJi6aM1sKqsY6isISJBcWZJLxVhZt2AW4BezrntCe1NzCwj+nwU8Qt/S51zq4EtZnZqNEtjIPBqWcfRyFlEwpLCac5m9gJwNtDYzPKBO4jPzqgLzIpmxH0Qzcw4C7jLzAqBIuAa59zei4mDic/8OJh4jTqxTl0ihbOIhCWWunR2zg0oofmPB1j3JeClA3w3Fzi+PMdWOItIWAK5Q1DhLCJhyVA4i4j4RyNnEREPhZHNCmcRCUwKLwimk8JZRMISRjYrnEUkLC4jjHvrFM4iEhaNnEVEPKTZGiIiHtIFQRERD4WRzQpnEQmMyhoiIh7S7dsiIh7SyDl8Tz5wNd3PPZFv122m83m/K/b9macey5Snb2bZirUAvDpzDvc88nKljlmnTi3++PC1nHhCW9Zv2Mql1z3C8vzv6Nwxi8fu/TUAZsaYh6cy7Y25lTqWpN/mzVu57bZH+fzzrzEzfv/7GzjxxGPS3a2aLYxsVjiX5k9T3uHJiW/w9MPXHnCd9+Ys4RdXPFDufR/RujFPPTSY8y+5+3vtl19yDhs2beP4s26kX8/TGDP8l1x23XgWfbaCMy4YQVHRHpo3bciHM+9l+uz5FBXtKfexxR9jxjzFmWeexPjxw9m9u4CdO3elu0s1ngtktkaZt9KY2TFmdouZjTezR6LPx1ZH59LtvY+WsH7j1gpt27/vT/jHtLv54PV7ePSeq4gl+Qfmgq4n8/zUdwF4ecaHnH1G/PncO3bu3hfEdevWxpX5ekjx3dat25kzZyEXXdQVgDp1alO/fr009yoAZskvHis1nM3sFmAS8X8ofATMiT6/YGbDqr57/jvlpHZ8OPNe/jrxFo7Njr/Dsf3RLbmo56mcc+GdnNp9OEVFjv59f5LU/lo2zyR/1ToAior2sHnLdg5vdBgAP+6UxbzZDzD37/dz/a1Pa9Rcw61Y8Q2ZmQ0YPnwcffrcwIgR49m+fWe6u1XzWTkWj5VV1rgK6OCcK0hsNLOxwCLg3qrqWE2wYOEy2p/2G7Zt38X553Ri8lM3ccJPb+KcM47npBOO4p+vjQbg4IPq8O26TQC8mHsTR7ZpQp06tWjTsjEfvH4PAI9PmMmfpryDlfC3efRyX+Ys+JKTf/ZftD+6JU+PHcwbb/+bXbsKiq0vNUNhYRGffvolI0deTceO7Rk9Opfc3KkMHXppurtWs/1Anq2xB2gJfL1fe4vouxKZWQ6QA1CrUWdq1Tu6Mn301patO/Z9fuOtBTwy+koOb3QYZsafp77L7fdNKrbNJTljgQPXnFeuXkfrloez8pv1ZGTEqH/YIcVKK5/lrWLb9l10aN+G+R8vrYIzk+rQvHljmjdvTMeO7QHo1u0McnOnprlXAfB8RJyssv6KGQq8aWavm1lutMwE3gRuONBGzrlc51xn51znUIMZoFmTBvs+d+6YRSxmrNuwhbfeW0jfHl1ocnh9ABo1OJQjWjVOap/TZ83jVxedBcCFPU7hnX8tAuDINk3IiEYER7RqTHZWS75e8W0qT0eqWZMmjWjevDFLl+YD8P77/yYrq02aexWAmCW/eKzUkbNzbqaZZQNdgFbE/07KB+Y454qqoX9pNfHR33DmacfSuNFh5H34GHePnUrt2vH/ZE//eTZ9e5zCf152HoWFRezcuZuBQ8YDsOSLlYx6cDKv/Xk4sViMgsJCbrztGZav/K7MYz774ttMGHctC999mA0bt3LZkEcBOP3H7bn52t4UFBSyZ4/jhhETWLdhS9WdvFSLkSOv5uabH6KgoJA2bZpxzz1D092lms/z0E2WuSq+7H/wEQM0r0CK2bF8VLq7IF7KrnSyHvXrKUlnztKn+3mb5JrnLCJh+YFcEBQRqVkCKWsonEUkLGEMnBXOIhIYz+/8S5bCWUTCorKGiIh/XCAj50CqMyIikVqW/FIGM5tgZmvNbGFCW6aZzTKzL6KfjaJ2ix4Ql2dmH5vZSQnbDIrW/8LMBiVzGgpnEQlLap9K9yzQbb+2YcCbzrl2xO+W3vsQuO5Au2jJAZ6Id8cygTuAU4jf0HfH3kAvjcJZRMKSwtu3nXPvAuv3a+4NTIw+TwT6JLQ/5+I+ABqaWQvgfGCWc269c24DMIvigV/8NJI6WRGRmqIcjww1sxwzm5uw5CRxhGbOudUA0c+mUXsrYEXCevlR24HaS6ULgiISlPK8CcU5lwvkpujQJR3YldJeKo2cRSQsVf9UujVRuYLo59qoPR9IfKxga2BVKe2ln0ZFeyci4qUMS36pmGnA3hkXg4BXE9oHRrM2TgU2RWWPN4CuZtYouhDYNWorlcoaIhKWFM5zNrMXgLOBxmaWT3zWxb3AZDO7ClgO9ItWnwH0APKA7cAVAM659WZ2N/HX/AHc5Zzb/yJjMQpnEQlLCu8QdM4NOMBX55awrgOuO8B+JgATynNshbOIhEW3b4uI+CeU27cVziISlopf6POKwllEwqKyhoiIhxTOIiIeCiObFc4iEpby3L7tM4WziIRFszVERDyk2RoiIv6JBfLEIIWziAQlkKqGwllEwqJwFhHxkAWSzgpnEQmKas4iIh4yhbOIiH8CqWoonEUkLIHcIKhwFpGwaOQsIuIhhbOIiIdiun1bRMQ/GjmLiHhI4Swi4iGFs4iIhzSVTkTEQxo5i4h4SLM1REQ8pJGziIiHQgnnQJ7fJCISZ5b8Uvp+rL2ZLUhYNpvZUDO708xWJrT3SNhmuJnlmdlnZnZ+Zc5DI2cRCUqqZms45z4DOgGYWQawEngFuAJ42Dn3YOL6ZnYc0B/oALQEZptZtnOuqCLH18hZRIISy0h+KYdzgS+dc1+Xsk5vYJJzbpdz7isgD+hS4fOo6IYiIj5KVVljP/2BFxJ+H2JmH5vZBDNrFLW1AlYkrJMftVWIwllEgmJm5VlyzGxuwpJTwv7qAL2AKVHTE0AW8ZLHauChvauW0B1X0fNQzVlEglKeEbFzLhfILWO17sB859yaaJs1/38sewr4W/RrPtAmYbvWwKrke/N9GjmLSFCqoKwxgISShpm1SPiuL7Aw+jwN6G9mdc2sLdAO+Kii51HlI+dtX4+s6kNIDbR+15J0d0E8lFk3u9L7SOU8ZzM7BDgPuDqh+X4z60S8ZLFs73fOuUVmNhn4FCgErqvoTA1QWUNEAlMrhfUA59x24PD92i4rZf0xwJhUHFvhLCJBiVmFr8F5ReEsIkHRI0NFRDwUyiwHhbOIBEVlDRERD6msISLioVoKZxER/5jKGiIi/lFZQ0TEQ5qtISLiIc3WEBHxkC4Iioh4SDVnEREPqawhIuIhjZxFRDyk2RoiIh5SWUNExEOpfNh+OimcRSQogWSzwllEwqKyhoiIhzRbQ0TEQypriIh4SCNnEREPZcRUcxYR8Y7KGiIiHtJsDRERD6nmLCLiIYWziIiHaqusISLin1BGzqFc2BQRAeLhnOxSFjNbZmafmNkCM5sbtWWa2Swz+yL62ShqNzMbb2Z5ZvaxmZ1UqfOozMYiIr7JsOSXJJ3jnOvknOsc/T4MeNM51w54M/odoDvQLlpygCcqcx4KZxEJSipHzgfQG5gYfZ4I9Elof87FfQA0NLMWFT6PCndPRMRDMXNJL2aWY2ZzE5ac/XbngL+b2byE75o551YDRD+bRu2tgBUJ2+ZHbRWiC4IiEpTa5RgRO+dygdxSVjnDObfKzJoCs8xsSSnrlnTkCk8d0chZRIKSyrKGc25V9HMt8ArQBVizt1wR/VwbrZ4PtEnYvDWwqsLnUdENRUR8VJ6yRmnM7FAzO2zvZ6ArsBCYBgyKVhsEvBp9ngYMjGZtnAps2lv+qAiVNUQkKOWYhVGWZsArZgbxrPyLc26mmc0BJpvZVcByoF+0/gygB5AHbAeuqMzBFc4iEpRU3YTinFsKdCyhfR1wbgntDrguNUdXOItIYPT2bRERD2Xo2RoiIv4JZOCscBaRsITy4COFs4gEReEsIuIh1ZxFRDyk2RoiIh5SWUNExEMpvEMwrRTOIhKUsp6ZUVMonKvI6tXfMeyWR/juuw1YLMbFF5/HwIE9eezRSUyZMovMzPoADL3xUn7605PT3Fspj9G3T+Zf73xKo8x6PP/KzcW+37x5O2Nun8zKFeuoU7c2I0ZdTFa75pU65u7dhdw1YhJLPs2nQYNDGP3ApbRolclH73/OH8bNoKCgiNq1Mxhy0wV0PuXoSh2rpguk5KxwrioZGTF+d8vldOiQxbatO/jFL37L6ad3AmDQoJ5ceVWfMvYgvvp5r8706386d42YVOL3E5/6H7Lbt+S+cZez7Ku1PDjmFR57+uqk9r165XruHvkif5gw+Hvtr738EYfVP5ip04cx6/UFPD5uBqMfuJQGDQ/lgUevoEnTBnz5xTcMHfwUr80eWelzrMlCqTmH8peMd5o2zaRDhywADq13MFlZrVmzZl2aeyWpcGLno6jf4JADfr9s6Ro6n9IOgB+1bco3q9azft0WAGb+bR5X/nI8A/uN5d67plJUtCepY/7j7UX06BX/F9Y5553A3A+/wDlH+2Nb0aRpAwCOOroZu3cVsnt3YWVOr8arHXNJLz6rcDibWaUeh/dDsjJ/LYsXf0XHjtkAPP/8DHr3GsqIWx9l06atae6dpNrR2S15+81PAFj0yXK+Wb2RtWs2sWzpGmbP/De5E6/juSk3kRGL8cb0+Unt89s1m2jWrCEAtWplUK/eQWzauP1767w16xOyj2lJnTo/7H8QV8M7BKtFZf4vjgKeKemL6F1bOQBPPHkHOTkXV+IwNdu2bTu4/vr7GDb8SurVO4T+A7ox+Np+mBnjH/kL99/3DGN+/5t0d1NSaOBV5/Dwfa8ysN9Ystq1IPuYlmRkxJjzYR6fLV7Jlb98BIBdOwtplFkPgFuGPsvqlespKChizeqNDOw3FoCLf3UmF/T5cYnvOrKEcFma9w1/GDedcf/9n1V9et7zPXSTVWo4m9nHB/qK+IOoS5T4Xq497lO//+1QhQoKCrnh+vvp2fMsunY9DYDGjRvu+75fv65cM3h0uronVeTQegdx292XAOCc48Lu99CyVSYL5i2le6+TufaGHsW2uW/c5cCBa85NmzVgzZqNNG3ekMLCIrZu3bmvtLL2m40Mu3EiI8f0p3WbxlV7cjVAKLXass6jGTAQ6FnCogJqKZxz3Hbb4xyV1ZrLr+i9r33t2vX7Ps+a/QHt2h2Zju5JFdqyeQcFBfG677SXPqLTSW05tN5BdD6lHW/N+oT16+KlrE2btrN61Yak9vmTs49jxrR5QLx8cXKXozEztmzewW+HTGDw9d3peGLbqjmhGsYs+cVnZZU1/gbUc84t2P8LM3u7SnoUiPnzFzPt1bfJzj6Svn1uBOLT5qZP/wdLFn+FmdGqVVPuHHVNmnsq5XX7755n/twv2bhxG71+NppfX9uVwsIiAC68+DSWfbWGu0a8SCxmtM1qxq2j4m8xapvVjKuHnM/Qa3LZs8dRq1YGN9/alxYtG5V5zJ59uzDq1klc9PN7qd/gEO6+/1cATJ30HvnLv+OZ3Nk8kzsbgHFP5pB5eL0qOnv/hVLWsPibVarOD7msIQe2cXdeursgHsqs26vS0Tr/u+lJZ85JjX/ubZT/sC/rikhwTHcIioj4x9uhcDkpnEUkKL5f6EuWwllEghJINiucRSQsemSoiIiHVNYQEfFQINmscBaRsCicRUQ8FModgqE8I0REBIiPnJNdSt2PWRsze8vMFpvZIjO7IWq/08xWmtmCaOmRsM1wM8szs8/M7PzKnIdGziISlBS+Q7AQ+K1zbr6ZHQbMM7NZ0XcPO+ceTFzZzI4D+gMdgJbAbDPLds4VVeTgGjmLSFBS9VQ659xq59z86PMWYDHQqpRNegOTnHO7nHNfAXlAl4qeh8JZRIISK8eSLDP7EXAi8GHUNMTMPjazCWa297GCrYAVCZvlU3qYl0rhLCJBKc/I2cxyzGxuwpJTfH9WD3gJGOqc2ww8AWQBnYDVwEN7Vy2hOxWusajmLCJBKc9kjcS3NpW4L7PaxIP5eefcy9E2axK+f4r4c+8hPlJuk7B5a2BVObrzPRo5i0hQUvWCVzMz4I/AYufc2IT2Fgmr9QUWRp+nAf3NrK6ZtQXaAR9V9Dw0chaRoKRwnvMZwGXAJ2a2921QtwIDzKwT8ZLFMuBqAOfcIjObDHxKfKbHdRWdqQEKZxEJTKqy2Tn3zwPsbkYp24wBxqTi+ApnEQmK3oQiIuKhQO7eVjiLSFj0yFAREQ9lpLsDKaJwFpGgaOQsIuKlMNJZ4SwiQTGFs4iIf8zCuPFZ4SwigdHIWUTEOxbII4MUziISFJU1RES8pLKGiIh3NFtDRMRDCmcREQ+ZhXEDt8JZRAKjkbOIiHdU1hAR8ZKm0omIeEcjZxERD1kgzwxVOItIUCyQx+0rnEUkMBo5i4h4R2UNEREvKZxFRLyjR4aKiHhJI2cREe/E9DxnEREfKZxFRLyjOwRFRLykcBYR8Y7mOYuIeCiU27fNOZfuPvxgmFmOcy433f0Qv+jPhZQkjMuaNUdOujsgXtKfCylG4Swi4iGFs4iIhxTO1Ut1RSmJ/lxIMbogKCLiIY2cRUQ8pHCuJmbWzcw+M7M8MxuW7v5I+pnZBDNba2YL090X8Y/CuRqYWQbwONAdOA4YYGbHpbdX4oFngW7p7oT4SeFcPboAec65pc653cAkoHea+yRp5px7F1if7n6InxTO1aMVsCLh9/yoTUSkRArn6lHSk1g0TUZEDkjhXD3ygTYJv7cGVqWpLyJSAyicq8ccoJ2ZtTWzOkB/YFqa+yQiHlM4VwPnXCEwBHgDWAxMds4tSm+vJN3M7AXgfaC9meWb2VXp7pP4Q3cIioh4SCNnEREPKZxFRDykcBYR8ZDCWUTEQwpnEREPKZxFRDykcBYR8ZDCWUTEQ/8H2e3Y7Fm5EUMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "y_pred=ensemble.predict(X_test_vect)\n",
    "CT_ensemble=confusion_matrix(y_test,y_pred)\n",
    "sns.heatmap(CT_SVM,annot=True,cmap='YlGnBu')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9947089947089947"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import precision_score\n",
    "precision_score(y_test,y_pred,pos_label='spam')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[2.66667048e-02, 9.73333295e-01],\n",
       "       [4.24293934e-07, 9.99999576e-01],\n",
       "       [9.99585772e-01, 4.14227898e-04],\n",
       "       [9.99712292e-01, 2.87708239e-04],\n",
       "       [9.99946988e-01, 5.30123039e-05],\n",
       "       [3.89312281e-01, 6.10687719e-01]])"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve, auc\n",
    "#Get the probability for each point in the test set.\n",
    "probs = ensemble.predict_proba(X_test_vect)\n",
    "probs[0:6]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "AUC score: 0.988\n"
     ]
    }
   ],
   "source": [
    "# Compute ROC curve and AUC for for the one class\n",
    "fpr, tpr, _ = roc_curve(y_test, probs[:,1],pos_label='spam')\n",
    "roc_auc = auc(fpr, tpr)\n",
    "  \n",
    "# Plot the ROC curve.\n",
    "plt.figure()\n",
    "lw = 2\n",
    "plt.plot(fpr, tpr, color='darkorange',\n",
    "         lw=lw, label='AUC = %0.2f' % roc_auc)\n",
    "plt.plot([0, 1], [0, 1], color='navy', lw=lw, linestyle='--')\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC and AUC')\n",
    "plt.legend(loc=\"best\")\n",
    "plt.show()\n",
    "print(\"AUC score: %3.3f\" % roc_auc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the heatmap and ROC curve its pretty evident that the ensemble model gives high precision =99.5% with minimal tuning"
   ]
  }
 ],
 "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}