SMS spam filter.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Building An SMS Spam Filter Using Multinomial Naive Bayes."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## Introduction:\n",
    "\n",
    "In this project, we are going to be using the `Multinomial Naive Bayes` algorithm to classify SMS as either spam messages or non-spam messages. The `Multinomial Naive Bayes` algorithm is based on `Bayes' Theorem`; this theorem describes the probability of an event based on prior knowledge of conditions that might be related to the event.\n",
    "\n",
    "Our goal in this project is to train the computer how to classify messages using the `Multinomial Naive Bayes` algorithm as well as a dataset of 5,572 SMS messages that have already been classified by humans. \n",
    "The dataset was put together by Tiago A. Almeida and José María Gómez Hidalgo and it can be downloaded from the <a href='https://archive.ics.uci.edu/ml/datasets/sms+spam+collection'>The UCI Machine Learning Repository<a/>. \n",
    "\n",
    "Our algorithm should at least have an accuracy score 80% for us to be confident in it.\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Initial Data Exploration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import re"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>Label</th>\n",
       "      <th>SMS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ham</td>\n",
       "      <td>Go until jurong point, crazy.. Available only ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ham</td>\n",
       "      <td>Ok lar... Joking wif u oni...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</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>3</th>\n",
       "      <td>ham</td>\n",
       "      <td>U dun say so early hor... U c already then say...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ham</td>\n",
       "      <td>Nah I don't think he goes to usf, he lives aro...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Label                                                SMS\n",
       "0   ham  Go until jurong point, crazy.. Available only ...\n",
       "1   ham                      Ok lar... Joking wif u oni...\n",
       "2  spam  Free entry in 2 a wkly comp to win FA Cup fina...\n",
       "3   ham  U dun say so early hor... U c already then say...\n",
       "4   ham  Nah I don't think he goes to usf, he lives aro..."
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sms = pd.read_csv('SMSSpamCollection',\n",
    "                 sep ='\\t',\n",
    "                 header = None,\n",
    "                 names = ['Label', 'SMS']\n",
    "                 )\n",
    "sms.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5572, 2)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sms.shape # displays the number of rows and columns in the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ham     86.593683\n",
       "spam    13.406317\n",
       "Name: Label, dtype: float64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sms['Label'].value_counts(normalize=True) * 100 # normalize = True displays the frequencies as proportions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Splitting Our Dataset Into Training and Test Datasets:\n",
    "\n",
    "Nearly 87% of the messages in our dataset are ham(non-spam) messages and about 13% of them are spam messages. Now that we are farmiliar with our dataset we are going to split it into a training set and a test set. \n",
    "We are going to keep 80% of our dataset for training and 20% for testing; this is because we want to train the algorithm on as much data as possible and still have enough data for testing the algorithm. The dataset has 5,572 messages, which means that:\n",
    "\n",
    "* The training set will have 4,458 messages (about 80% of the dataset).\n",
    "* The test set will have 1,114 messages (about 20% of the dataset)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "sample = sms.sample(frac=1, random_state=1) # randomizes the dataset\n",
    "training_set = sample.iloc[0:4458].reset_index(drop=True) # (drop=True) drops old index after resetting index \n",
    "test_set = sample.iloc[4458:].reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(4458, 2)\n",
      "(1114, 2)\n"
     ]
    }
   ],
   "source": [
    "print(training_set.shape)\n",
    "\n",
    "print(test_set.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "training set: \n",
      " ham     86.54105\n",
      "spam    13.45895\n",
      "Name: Label, dtype: float64\n",
      "\n",
      "\n",
      "test set: \n",
      " ham     86.804309\n",
      "spam    13.195691\n",
      "Name: Label, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "#getting the percentages of spam and non-spam messages in our training and tests set\n",
    "print('training set:', '\\n',\n",
    "      training_set['Label'].value_counts(normalize=True) * 100\n",
    "     )\n",
    "print('\\n')\n",
    "print('test set:', '\\n',\n",
    "      test_set['Label'].value_counts(normalize=True) * 100\n",
    "     )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The perecentages for spam and non-spam messages in both our training and test dataset are roughly equal to the percentages of spam and non-spam messages in our original sms dataset."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Cleaning.\n",
    "To train the algorithm we are going to clean our dataset t make it easier for to calculate the probability of each individual word in the dataset. We are going to transform the dataset. The result we will get is going to look like this.\n",
    "\n",
    "<img src = 'https://dq-content.s3.amazonaws.com/433/cpgp_dataset_3.png'/>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>Label</th>\n",
       "      <th>SMS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ham</td>\n",
       "      <td>Yep, by the pretty sculpture</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ham</td>\n",
       "      <td>Yes, princess. Are you going to make me moan?</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ham</td>\n",
       "      <td>Welp apparently he retired</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ham</td>\n",
       "      <td>Havent.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ham</td>\n",
       "      <td>I forgot 2 ask ü all smth.. There's a card on ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Label                                                SMS\n",
       "0   ham                       Yep, by the pretty sculpture\n",
       "1   ham      Yes, princess. Are you going to make me moan?\n",
       "2   ham                         Welp apparently he retired\n",
       "3   ham                                            Havent.\n",
       "4   ham  I forgot 2 ask ü all smth.. There's a card on ..."
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_set.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "#removing punctuations\n",
    "training_set['SMS'] = training_set['SMS'].str.replace('\\W', ' ', regex=True)\n",
    "\n",
    "#turning every text to lower case\n",
    "training_set['SMS'] = training_set['SMS'].str.lower()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Label</th>\n",
       "      <th>SMS</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ham</td>\n",
       "      <td>yep  by the pretty sculpture</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ham</td>\n",
       "      <td>yes  princess  are you going to make me moan</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ham</td>\n",
       "      <td>welp apparently he retired</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ham</td>\n",
       "      <td>havent</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ham</td>\n",
       "      <td>i forgot 2 ask ü all smth   there s a card on ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Label                                                SMS\n",
       "0   ham                       yep  by the pretty sculpture\n",
       "1   ham      yes  princess  are you going to make me moan \n",
       "2   ham                         welp apparently he retired\n",
       "3   ham                                            havent \n",
       "4   ham  i forgot 2 ask ü all smth   there s a card on ..."
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "training_set.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "7783"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#extracting every individual word in the sms column\n",
    "training_set['SMS'] = training_set['SMS'].str.split()\n",
    "vocabulary = []\n",
    "for value in training_set['SMS']:\n",
    "    for i in value:\n",
    "        vocabulary.append(i)\n",
    "vocabulary = set(vocabulary) # using the set function gets rid of duplicate values\n",
    "vocabulary = list(vocabulary) \n",
    "\n",
    "len(vocabulary)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Final Transformation Of The Training Dataset.\n",
    "\n",
    "There are 7783 unique words in our vocabulary. We are going to create a dictionary containing each unique word as a key and the frequency of the word as its key. Then we are going to transform the dictionary to a pandas dataframe and then combine it with our `training_set` dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "#creates a dictionary with every unique word in the vocabulary list\n",
    "word_counts_per_sms = {unique_word: [0] * len(training_set['SMS']) for unique_word in vocabulary}\n",
    "\n",
    "for index, message in enumerate(training_set['SMS']):\n",
    "    for word in message:\n",
    "        word_counts_per_sms[word][index] += 1"
   ]
  },
  {
   "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>buns</th>\n",
       "      <th>nigro</th>\n",
       "      <th>restrict</th>\n",
       "      <th>pockets</th>\n",
       "      <th>awesome</th>\n",
       "      <th>hair</th>\n",
       "      <th>sportsx</th>\n",
       "      <th>mittelschmertz</th>\n",
       "      <th>formal</th>\n",
       "      <th>84199</th>\n",
       "      <th>...</th>\n",
       "      <th>light</th>\n",
       "      <th>dey</th>\n",
       "      <th>messages</th>\n",
       "      <th>fuckin</th>\n",
       "      <th>teams</th>\n",
       "      <th>babysit</th>\n",
       "      <th>almost</th>\n",
       "      <th>page</th>\n",
       "      <th>access</th>\n",
       "      <th>passed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 7783 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   buns  nigro  restrict  pockets  awesome  hair  sportsx  mittelschmertz  \\\n",
       "0     0      0         0        0        0     0        0               0   \n",
       "1     0      0         0        0        0     0        0               0   \n",
       "2     0      0         0        0        0     0        0               0   \n",
       "3     0      0         0        0        0     0        0               0   \n",
       "4     0      0         0        0        0     0        0               0   \n",
       "\n",
       "   formal  84199  ...  light  dey  messages  fuckin  teams  babysit  almost  \\\n",
       "0       0      0  ...      0    0         0       0      0        0       0   \n",
       "1       0      0  ...      0    0         0       0      0        0       0   \n",
       "2       0      0  ...      0    0         0       0      0        0       0   \n",
       "3       0      0  ...      0    0         0       0      0        0       0   \n",
       "4       0      0  ...      0    0         0       0      0        0       0   \n",
       "\n",
       "   page  access  passed  \n",
       "0     0       0       0  \n",
       "1     0       0       0  \n",
       "2     0       0       0  \n",
       "3     0       0       0  \n",
       "4     0       0       0  \n",
       "\n",
       "[5 rows x 7783 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "word_counts = pd.DataFrame(word_counts_per_sms)\n",
    "word_counts.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>Label</th>\n",
       "      <th>SMS</th>\n",
       "      <th>buns</th>\n",
       "      <th>nigro</th>\n",
       "      <th>restrict</th>\n",
       "      <th>pockets</th>\n",
       "      <th>awesome</th>\n",
       "      <th>hair</th>\n",
       "      <th>sportsx</th>\n",
       "      <th>mittelschmertz</th>\n",
       "      <th>...</th>\n",
       "      <th>light</th>\n",
       "      <th>dey</th>\n",
       "      <th>messages</th>\n",
       "      <th>fuckin</th>\n",
       "      <th>teams</th>\n",
       "      <th>babysit</th>\n",
       "      <th>almost</th>\n",
       "      <th>page</th>\n",
       "      <th>access</th>\n",
       "      <th>passed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ham</td>\n",
       "      <td>[yep, by, the, pretty, sculpture]</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ham</td>\n",
       "      <td>[yes, princess, are, you, going, to, make, me,...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>ham</td>\n",
       "      <td>[welp, apparently, he, retired]</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ham</td>\n",
       "      <td>[havent]</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ham</td>\n",
       "      <td>[i, forgot, 2, ask, ü, all, smth, there, s, a,...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 7785 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  Label                                                SMS  buns  nigro  \\\n",
       "0   ham                  [yep, by, the, pretty, sculpture]     0      0   \n",
       "1   ham  [yes, princess, are, you, going, to, make, me,...     0      0   \n",
       "2   ham                    [welp, apparently, he, retired]     0      0   \n",
       "3   ham                                           [havent]     0      0   \n",
       "4   ham  [i, forgot, 2, ask, ü, all, smth, there, s, a,...     0      0   \n",
       "\n",
       "   restrict  pockets  awesome  hair  sportsx  mittelschmertz  ...  light  dey  \\\n",
       "0         0        0        0     0        0               0  ...      0    0   \n",
       "1         0        0        0     0        0               0  ...      0    0   \n",
       "2         0        0        0     0        0               0  ...      0    0   \n",
       "3         0        0        0     0        0               0  ...      0    0   \n",
       "4         0        0        0     0        0               0  ...      0    0   \n",
       "\n",
       "   messages  fuckin  teams  babysit  almost  page  access  passed  \n",
       "0         0       0      0        0       0     0       0       0  \n",
       "1         0       0      0        0       0     0       0       0  \n",
       "2         0       0      0        0       0     0       0       0  \n",
       "3         0       0      0        0       0     0       0       0  \n",
       "4         0       0      0        0       0     0       0       0  \n",
       "\n",
       "[5 rows x 7785 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clean_training_set = pd.concat([training_set, word_counts], axis=1) # joins the dataset together\n",
    "clean_training_set.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Probability Of Spam & Non-Spam Messages.\n",
    "Here, we are going to do the following:\n",
    "\n",
    "* split our training dataset into datasets for ham and spam messages.\n",
    "* calculate the probability of the constants. To do this, we are going:\n",
    "  1. Calculate the probability of both ham and spam messages.\n",
    "  2. Create a variable alpha with a value of 1.\n",
    "  3. Calculate the number of words in both ham and spam messages.\n",
    "  4. Calculate the number of words in the vocabulary dictionary.\n",
    "  \n",
    "* Calculate the parameters. This includes the probability of every word given that a message is either spam or ham. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Calculating the constants:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "probability of spam messages: \n",
      " 0.13458950201884254\n",
      "\n",
      "\n",
      "probability of non-spam messages: \n",
      " 0.8654104979811574\n"
     ]
    }
   ],
   "source": [
    "spam_messages = clean_training_set[clean_training_set['Label']=='spam'].shape[0] # returns the number of rows with spam messages\n",
    "\n",
    "ham_messages = clean_training_set[clean_training_set['Label']=='ham'].shape[0] # returns the number of rows with ham messages\n",
    "\n",
    "total_messages = clean_training_set.shape[0]\n",
    "\n",
    "probability_spam = spam_messages / total_messages\n",
    "probability_ham = ham_messages / total_messages\n",
    "\n",
    "print('probability of spam messages:', '\\n', probability_spam)\n",
    "print('\\n')\n",
    "print('probability of non-spam messages:', '\\n', probability_ham)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "spam_words = clean_training_set[clean_training_set['Label']=='spam']['SMS'].apply(len) # len function counts the individual words in the sms column\n",
    "n_spam = spam_words.sum() # sums them to get the total amount of spam words in the SMS column\n",
    "\n",
    "ham_words = clean_training_set[clean_training_set['Label']=='ham']['SMS'].apply(len)\n",
    "n_ham = ham_words.sum()\n",
    "\n",
    "n_vocabulary = len(vocabulary)\n",
    "alpha = 1 #laplace smoothing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Calculating the parameters:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "spam_dict = {unique_word:0 for unique_word in vocabulary} # initializes  a dictionary with every word in the vocabulary list as a key.\n",
    "ham_dict = {unique_word:0 for unique_word in vocabulary}\n",
    "\n",
    "# creating new DataFrames for both Spam and Ham messages\n",
    "spam_df = clean_training_set[clean_training_set['Label']=='spam'].copy() \n",
    "ham_df = clean_training_set[clean_training_set['Label']=='ham'].copy()\n",
    "\n",
    "#calculating the probability for each word in spam messages\n",
    "for word in vocabulary:\n",
    "    n_word_spam_messages = spam_df[word].sum() # the number of times the word occurs in the spam DataFrame\n",
    "    p_word_spam_messages = (n_word_spam_messages + alpha) / (n_spam + alpha * n_vocabulary)\n",
    "    spam_dict[word] =  p_word_spam_messages # updates the dictionary values with the probability of each unique word\n",
    "    \n",
    "\n",
    "#calculating the probability for each word in ham messages\n",
    "for word in vocabulary:\n",
    "    n_word_ham_messages = ham_df[word].sum() # the number of times the word occurs in the spam DataFrame\n",
    "    p_word_ham_messages = (n_word_ham_messages + alpha) / (n_ham + alpha * n_vocabulary)\n",
    "    ham_dict[word] =  p_word_ham_messages # updates the dictionary values with the probability of each unique word\n",
    "        \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Classifying Messages:\n",
    "\n",
    "* First we are going to write a mock `classify()`function that does the classification and confirm that it can classify messages as spam or non-spam.\n",
    "* We are going to update the `classify()` function and then use it on our test_set DataFrame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def classify(message):\n",
    "    \"\"\"function to classify sms as spam or non-spam messages\"\"\"\n",
    "\n",
    "    message = re.sub('\\W', ' ', message) # replaces any special character with an empty string\n",
    "    message = message.lower()\n",
    "    message = message.split()\n",
    "\n",
    "\n",
    "    p_spam_given_message = probability_spam\n",
    "    p_ham_given_message = probability_ham\n",
    "    \n",
    "    for word in message: # checks if the word is in ham_dict and spam_dict and gets their probability\n",
    "        if word in spam_dict:\n",
    "            p_spam_given_message *= spam_dict[word] \n",
    "\n",
    "            \n",
    "        if word in ham_dict:\n",
    "            p_ham_given_message *= ham_dict[word]\n",
    "            \n",
    "   \n",
    "\n",
    "    if p_ham_given_message > p_spam_given_message:\n",
    "        return 'This is not a spam message.'\n",
    "    elif p_ham_given_message < p_spam_given_message:\n",
    "        return 'This is a spam message.'\n",
    "    else:\n",
    "        return 'Unsure, have a human classify this!'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This is a spam message.\n",
      "\n",
      "\n",
      "This is not a spam message.\n"
     ]
    }
   ],
   "source": [
    "# Testing the fucntion\n",
    "test_message1 = 'WINNER!! This is the secret code to unlock the money: C3421.'\n",
    "test_message2 = \"Sounds good, Tom, then see u there\"\n",
    "\n",
    "print(classify(test_message1))\n",
    "print('\\n')\n",
    "print(classify(test_message2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "def classify_test_set(message):\n",
    "    \"\"\"function to classify messages as ham or spam\"\"\"\n",
    "\n",
    "    message = re.sub('\\W', ' ', message) # replaces any special character with an empty string\n",
    "    message = message.lower()\n",
    "    message = message.split()\n",
    "\n",
    "\n",
    "    p_spam_given_message = probability_spam\n",
    "    p_ham_given_message = probability_ham\n",
    "    \n",
    "    for word in message: # checks if the word is in ham_dict and spam_dict and gets their probability\n",
    "        if word in spam_dict:\n",
    "            p_spam_given_message *= spam_dict[word] \n",
    "            \n",
    "            \n",
    "        if word in ham_dict:\n",
    "            p_ham_given_message *= ham_dict[word]\n",
    "            \n",
    "\n",
    "    if p_ham_given_message > p_spam_given_message:\n",
    "        return 'ham'\n",
    "    elif p_ham_given_message < p_spam_given_message:\n",
    "        return 'spam'\n",
    "    else:\n",
    "        return 'needs human classification'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Label</th>\n",
       "      <th>SMS</th>\n",
       "      <th>predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ham</td>\n",
       "      <td>Later i guess. I needa do mcat study too.</td>\n",
       "      <td>ham</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ham</td>\n",
       "      <td>But i haf enuff space got like 4 mb...</td>\n",
       "      <td>ham</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>spam</td>\n",
       "      <td>Had your mobile 10 mths? Update to latest Oran...</td>\n",
       "      <td>spam</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ham</td>\n",
       "      <td>All sounds good. Fingers . Makes it difficult ...</td>\n",
       "      <td>ham</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ham</td>\n",
       "      <td>All done, all handed in. Don't know if mega sh...</td>\n",
       "      <td>ham</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Label                                                SMS predicted\n",
       "0   ham          Later i guess. I needa do mcat study too.       ham\n",
       "1   ham             But i haf enuff space got like 4 mb...       ham\n",
       "2  spam  Had your mobile 10 mths? Update to latest Oran...      spam\n",
       "3   ham  All sounds good. Fingers . Makes it difficult ...       ham\n",
       "4   ham  All done, all handed in. Don't know if mega sh...       ham"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#applying the classify function to the training_set DataFrame\n",
    "test_set['predicted'] = test_set['SMS'].apply(classify_test_set)\n",
    "test_set.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calculating The Accuracy Of The Algorithm.\n",
    "To calculate the accuracy of the algorithm, we are going to do the following:\n",
    "\n",
    "* initialize a variable `correct` with a value of 0 and also a variable `total` with the total number of messages in the test set.\n",
    "* Loop through our `test_set` DataFrame and increment the value of correct by 1 if the `Label` column is the same as the `predicted` column.\n",
    "* Finally we divide `correct` by `total` to get our accuracy score."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# function labels each row as correct or incorrect if we predicted correctly\n",
    "def classification(row):\n",
    "    if row['Label'] != row['predicted']:\n",
    "        return 'incorrect'\n",
    "    else:\n",
    "        return 'correct'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": true
   },
   "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>Label</th>\n",
       "      <th>SMS</th>\n",
       "      <th>predicted</th>\n",
       "      <th>classification</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ham</td>\n",
       "      <td>Later i guess. I needa do mcat study too.</td>\n",
       "      <td>ham</td>\n",
       "      <td>correct</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>ham</td>\n",
       "      <td>But i haf enuff space got like 4 mb...</td>\n",
       "      <td>ham</td>\n",
       "      <td>correct</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>spam</td>\n",
       "      <td>Had your mobile 10 mths? Update to latest Oran...</td>\n",
       "      <td>spam</td>\n",
       "      <td>correct</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>ham</td>\n",
       "      <td>All sounds good. Fingers . Makes it difficult ...</td>\n",
       "      <td>ham</td>\n",
       "      <td>correct</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>ham</td>\n",
       "      <td>All done, all handed in. Don't know if mega sh...</td>\n",
       "      <td>ham</td>\n",
       "      <td>correct</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Label                                                SMS predicted  \\\n",
       "0   ham          Later i guess. I needa do mcat study too.       ham   \n",
       "1   ham             But i haf enuff space got like 4 mb...       ham   \n",
       "2  spam  Had your mobile 10 mths? Update to latest Oran...      spam   \n",
       "3   ham  All sounds good. Fingers . Makes it difficult ...       ham   \n",
       "4   ham  All done, all handed in. Don't know if mega sh...       ham   \n",
       "\n",
       "  classification  \n",
       "0        correct  \n",
       "1        correct  \n",
       "2        correct  \n",
       "3        correct  \n",
       "4        correct  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# creates a new colun that shows the classification of each column\n",
    "test_set['classification'] = test_set.apply(classification, axis=1)\n",
    "\n",
    "test_set.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": true
   },
   "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>Label</th>\n",
       "      <th>SMS</th>\n",
       "      <th>predicted</th>\n",
       "      <th>classification</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>114</th>\n",
       "      <td>spam</td>\n",
       "      <td>Not heard from U4 a while. Call me now am here...</td>\n",
       "      <td>ham</td>\n",
       "      <td>incorrect</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>spam</td>\n",
       "      <td>More people are dogging in your area now. Call...</td>\n",
       "      <td>ham</td>\n",
       "      <td>incorrect</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>152</th>\n",
       "      <td>ham</td>\n",
       "      <td>Unlimited texts. Limited minutes.</td>\n",
       "      <td>spam</td>\n",
       "      <td>incorrect</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>159</th>\n",
       "      <td>ham</td>\n",
       "      <td>26th OF JULY</td>\n",
       "      <td>spam</td>\n",
       "      <td>incorrect</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>284</th>\n",
       "      <td>ham</td>\n",
       "      <td>Nokia phone is lovly..</td>\n",
       "      <td>spam</td>\n",
       "      <td>incorrect</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>293</th>\n",
       "      <td>ham</td>\n",
       "      <td>A Boy loved a gal. He propsd bt she didnt mind...</td>\n",
       "      <td>needs human classification</td>\n",
       "      <td>incorrect</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>302</th>\n",
       "      <td>ham</td>\n",
       "      <td>No calls..messages..missed calls</td>\n",
       "      <td>spam</td>\n",
       "      <td>incorrect</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>319</th>\n",
       "      <td>ham</td>\n",
       "      <td>We have sent JD for Customer Service cum Accou...</td>\n",
       "      <td>spam</td>\n",
       "      <td>incorrect</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>504</th>\n",
       "      <td>spam</td>\n",
       "      <td>Oh my god! I've found your number again! I'm s...</td>\n",
       "      <td>ham</td>\n",
       "      <td>incorrect</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>546</th>\n",
       "      <td>spam</td>\n",
       "      <td>Hi babe its Chloe, how r u? I was smashed on s...</td>\n",
       "      <td>ham</td>\n",
       "      <td>incorrect</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>741</th>\n",
       "      <td>spam</td>\n",
       "      <td>0A$NETWORKS allow companies to bill for SMS, s...</td>\n",
       "      <td>ham</td>\n",
       "      <td>incorrect</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>876</th>\n",
       "      <td>spam</td>\n",
       "      <td>RCT' THNQ Adrian for U text. Rgds Vatian</td>\n",
       "      <td>ham</td>\n",
       "      <td>incorrect</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>885</th>\n",
       "      <td>spam</td>\n",
       "      <td>2/2 146tf150p</td>\n",
       "      <td>ham</td>\n",
       "      <td>incorrect</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>953</th>\n",
       "      <td>spam</td>\n",
       "      <td>Hello. We need some posh birds and chaps to us...</td>\n",
       "      <td>ham</td>\n",
       "      <td>incorrect</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Label                                                SMS  \\\n",
       "114  spam  Not heard from U4 a while. Call me now am here...   \n",
       "135  spam  More people are dogging in your area now. Call...   \n",
       "152   ham                  Unlimited texts. Limited minutes.   \n",
       "159   ham                                       26th OF JULY   \n",
       "284   ham                             Nokia phone is lovly..   \n",
       "293   ham  A Boy loved a gal. He propsd bt she didnt mind...   \n",
       "302   ham                   No calls..messages..missed calls   \n",
       "319   ham  We have sent JD for Customer Service cum Accou...   \n",
       "504  spam  Oh my god! I've found your number again! I'm s...   \n",
       "546  spam  Hi babe its Chloe, how r u? I was smashed on s...   \n",
       "741  spam  0A$NETWORKS allow companies to bill for SMS, s...   \n",
       "876  spam           RCT' THNQ Adrian for U text. Rgds Vatian   \n",
       "885  spam                                      2/2 146tf150p   \n",
       "953  spam  Hello. We need some posh birds and chaps to us...   \n",
       "\n",
       "                      predicted classification  \n",
       "114                         ham      incorrect  \n",
       "135                         ham      incorrect  \n",
       "152                        spam      incorrect  \n",
       "159                        spam      incorrect  \n",
       "284                        spam      incorrect  \n",
       "293  needs human classification      incorrect  \n",
       "302                        spam      incorrect  \n",
       "319                        spam      incorrect  \n",
       "504                         ham      incorrect  \n",
       "546                         ham      incorrect  \n",
       "741                         ham      incorrect  \n",
       "876                         ham      incorrect  \n",
       "885                         ham      incorrect  \n",
       "953                         ham      incorrect  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# calculating the number of incorrect columns\n",
    "incorrect = test_set[test_set['classification'] == 'incorrect']\n",
    "incorrect"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "13"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# getting the number of incorrectly classified messages\n",
    "incorrect.shape[0] - 1 # subtract one to account for the message that needs human classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "98.74326750448833"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# calculating the accuracy of the algorithm\n",
    "correct = test_set[test_set['classification'] == 'correct'].shape[0]\n",
    "total = test_set.shape[0]\n",
    "\n",
    "accuracy = correct / total * 100\n",
    "accuracy"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the accuracy score, our algorithm is doing pretty good. We managed to classify 98.7% of messages correctly."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion:\n",
    "We wanted to create an algorithm that classifies messages as either spam or non-spam messages using the multinomial Naive Bayes algorithm. So far we have been able to do that and we had an accuracy score of 98.7%. This means that we are confident that in the real world our algorithm will do so well.\n",
    "Although our algorithm is doing quite good, it failed to classify 13 messsages correctly. There are a few things we can do to improve on it. \n",
    "1. We could make the algorithm case sensitive to see if it will increase the accuracy.\n",
    "2. We could calculate the probability for certain phrases appearin in spam messages instead od just individual words.\n",
    "3. Collect more data on sms classification and retrain the model."
   ]
  }
 ],
 "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}