CA0.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduction to AI\n",
    "## Ghazal Kalhor"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div style=\"text-align: justify; font-weight: bold;\">\n",
    "<i>Abstract</i> — In this computer assignment, we want to learn how to\n",
    "work with some of the most important libraries in Python such as\n",
    "NumPy, Pandas, and matplotlib that will help us in learning\n",
    "Artificial Intelligence; this is done by working with data frames\n",
    "and using them to predict a simple statistical model.  \n",
    "    \n",
    "<i>Keywords</i> — Python, NumPy, Pandas, Matplotlib, Artificial\n",
    "Intelligence\n",
    " </div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Introduction\n",
    "<br/>\n",
    "<div style=\"text-align: justify;\"> \n",
    "The goal of this computer assignment is to design a simple statistical\n",
    "model in order to predict missing values of the target column; this is done by using samples that have a pre-determined target value.\n",
    "Some characteristics of users such as Age, Area Income, and Gender are given in this computer assignment, The input of our model will be one of these characteristics and the output is the target column which determines whether the user has clicked on the Ad or not.\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Importing Libraries\n",
    "<br/>\n",
    "<div style=\"text-align: justify;\">\n",
    "In this part, some of the necessary libraries were imported in order to use their helpful functions. These libraries are mostly used in data science applications.\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from time import time\n",
    "from scipy.stats import norm\n",
    "import matplotlib.pyplot as plt\n",
    "from pandas.api.types import is_numeric_dtype"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Defining Constants\n",
    "<br/>\n",
    "<div style=\"text-align: justify;\">\n",
    "In this part constant values are defined in order to make the code more readable and more flexible to change.\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "TRUE = 1\n",
    "FALSE = 0\n",
    "MALE = 1\n",
    "FEMALE = 0\n",
    "MY_AGE = 21\n",
    "CLICKED = 1\n",
    "NOT_CLICKED = 0\n",
    "TARGET = 'Clicked on Ad'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Importing Data\n",
    "\n",
    "Contents of file *advertising_dataset.csv* was read using *pd.read_csv* and then stored in the *advertising* dataframe.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "advertising = pd.read_csv('advertising_dataset.csv', index_col = 'Unnamed: 0')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pandas *tail* method is used in order to return the last _n_ rows of the dataframe. *n* is 5 by default."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Daily Time Spent on Site</th>\n",
       "      <th>Age</th>\n",
       "      <th>Area Income</th>\n",
       "      <th>Daily Internet Usage</th>\n",
       "      <th>Ad Topic Line</th>\n",
       "      <th>City</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Country</th>\n",
       "      <th>Timestamp</th>\n",
       "      <th>Clicked on Ad</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>995</th>\n",
       "      <td>72.97</td>\n",
       "      <td>30.0</td>\n",
       "      <td>71384.57</td>\n",
       "      <td>208.58</td>\n",
       "      <td>Fundamental modular algorithm</td>\n",
       "      <td>Duffystad</td>\n",
       "      <td>Male</td>\n",
       "      <td>Lebanon</td>\n",
       "      <td>2016-02-11 21:49:00</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>51.30</td>\n",
       "      <td>45.0</td>\n",
       "      <td>67782.17</td>\n",
       "      <td>134.42</td>\n",
       "      <td>Grass-roots cohesive monitoring</td>\n",
       "      <td>New Darlene</td>\n",
       "      <td>Male</td>\n",
       "      <td>Bosnia and Herzegovina</td>\n",
       "      <td>2016-04-22 02:07:01</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>51.63</td>\n",
       "      <td>51.0</td>\n",
       "      <td>42415.72</td>\n",
       "      <td>120.37</td>\n",
       "      <td>Expanded intangible solution</td>\n",
       "      <td>South Jessica</td>\n",
       "      <td>Male</td>\n",
       "      <td>Mongolia</td>\n",
       "      <td>2016-02-01 17:24:57</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>55.55</td>\n",
       "      <td>19.0</td>\n",
       "      <td>41920.79</td>\n",
       "      <td>187.95</td>\n",
       "      <td>Proactive bandwidth-monitored policy</td>\n",
       "      <td>West Steven</td>\n",
       "      <td>Female</td>\n",
       "      <td>Guatemala</td>\n",
       "      <td>2016-03-24 02:35:54</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>999</th>\n",
       "      <td>NaN</td>\n",
       "      <td>26.0</td>\n",
       "      <td>29875.80</td>\n",
       "      <td>178.35</td>\n",
       "      <td>Virtual 5thgeneration emulation</td>\n",
       "      <td>Ronniemouth</td>\n",
       "      <td>Female</td>\n",
       "      <td>Brazil</td>\n",
       "      <td>2016-06-03 21:43:21</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Daily Time Spent on Site   Age  Area Income  Daily Internet Usage  \\\n",
       "995                     72.97  30.0     71384.57                208.58   \n",
       "996                     51.30  45.0     67782.17                134.42   \n",
       "997                     51.63  51.0     42415.72                120.37   \n",
       "998                     55.55  19.0     41920.79                187.95   \n",
       "999                       NaN  26.0     29875.80                178.35   \n",
       "\n",
       "                            Ad Topic Line           City  Gender  \\\n",
       "995         Fundamental modular algorithm      Duffystad    Male   \n",
       "996       Grass-roots cohesive monitoring    New Darlene    Male   \n",
       "997          Expanded intangible solution  South Jessica    Male   \n",
       "998  Proactive bandwidth-monitored policy    West Steven  Female   \n",
       "999       Virtual 5thgeneration emulation    Ronniemouth  Female   \n",
       "\n",
       "                    Country            Timestamp  Clicked on Ad  \n",
       "995                 Lebanon  2016-02-11 21:49:00            1.0  \n",
       "996  Bosnia and Herzegovina  2016-04-22 02:07:01            1.0  \n",
       "997                Mongolia  2016-02-01 17:24:57            1.0  \n",
       "998               Guatemala  2016-03-24 02:35:54            0.0  \n",
       "999                  Brazil  2016-06-03 21:43:21            1.0  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "advertising.tail()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pandas *head* method is used in order to return the first _n_ rows of the dataframe. *n* is 5 by default."
   ]
  },
  {
   "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>Daily Time Spent on Site</th>\n",
       "      <th>Age</th>\n",
       "      <th>Area Income</th>\n",
       "      <th>Daily Internet Usage</th>\n",
       "      <th>Ad Topic Line</th>\n",
       "      <th>City</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Country</th>\n",
       "      <th>Timestamp</th>\n",
       "      <th>Clicked on Ad</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>68.95</td>\n",
       "      <td>35.0</td>\n",
       "      <td>61833.90</td>\n",
       "      <td>256.09</td>\n",
       "      <td>Cloned 5thgeneration orchestration</td>\n",
       "      <td>Wrightburgh</td>\n",
       "      <td>Female</td>\n",
       "      <td>Tunisia</td>\n",
       "      <td>2016-03-27 00:53:11</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>80.23</td>\n",
       "      <td>31.0</td>\n",
       "      <td>68441.85</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Monitored national standardization</td>\n",
       "      <td>West Jodi</td>\n",
       "      <td>Male</td>\n",
       "      <td>Nauru</td>\n",
       "      <td>2016-04-04 01:39:02</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>69.47</td>\n",
       "      <td>26.0</td>\n",
       "      <td>59785.94</td>\n",
       "      <td>236.50</td>\n",
       "      <td>Organic bottom-line service-desk</td>\n",
       "      <td>Davidton</td>\n",
       "      <td>Female</td>\n",
       "      <td>San Marino</td>\n",
       "      <td>2016-03-13 20:35:42</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>74.15</td>\n",
       "      <td>29.0</td>\n",
       "      <td>54806.18</td>\n",
       "      <td>245.89</td>\n",
       "      <td>Triple-buffered reciprocal time-frame</td>\n",
       "      <td>West Terrifurt</td>\n",
       "      <td>Male</td>\n",
       "      <td>Italy</td>\n",
       "      <td>2016-01-10 02:31:19</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>68.37</td>\n",
       "      <td>35.0</td>\n",
       "      <td>73889.99</td>\n",
       "      <td>225.58</td>\n",
       "      <td>Robust logistical utilization</td>\n",
       "      <td>South Manuel</td>\n",
       "      <td>Female</td>\n",
       "      <td>Iceland</td>\n",
       "      <td>2016-06-03 03:36:18</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Daily Time Spent on Site   Age  Area Income  Daily Internet Usage  \\\n",
       "0                     68.95  35.0     61833.90                256.09   \n",
       "1                     80.23  31.0     68441.85                   NaN   \n",
       "2                     69.47  26.0     59785.94                236.50   \n",
       "3                     74.15  29.0     54806.18                245.89   \n",
       "4                     68.37  35.0     73889.99                225.58   \n",
       "\n",
       "                           Ad Topic Line            City  Gender     Country  \\\n",
       "0     Cloned 5thgeneration orchestration     Wrightburgh  Female     Tunisia   \n",
       "1     Monitored national standardization       West Jodi    Male       Nauru   \n",
       "2       Organic bottom-line service-desk        Davidton  Female  San Marino   \n",
       "3  Triple-buffered reciprocal time-frame  West Terrifurt    Male       Italy   \n",
       "4          Robust logistical utilization    South Manuel  Female     Iceland   \n",
       "\n",
       "             Timestamp  Clicked on Ad  \n",
       "0  2016-03-27 00:53:11            0.0  \n",
       "1  2016-04-04 01:39:02            0.0  \n",
       "2  2016-03-13 20:35:42            0.0  \n",
       "3  2016-01-10 02:31:19            0.0  \n",
       "4  2016-06-03 03:36:18            0.0  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "advertising.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Basic Statistical Details of  Data\n",
    "Pandas *describe* method is used in order to view some basic statistical details like max, min, percentile, mean, std etc. of the dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>Daily Time Spent on Site</th>\n",
       "      <th>Age</th>\n",
       "      <th>Area Income</th>\n",
       "      <th>Daily Internet Usage</th>\n",
       "      <th>Clicked on Ad</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>907.000000</td>\n",
       "      <td>979.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>913.000000</td>\n",
       "      <td>832.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>64.794818</td>\n",
       "      <td>36.009193</td>\n",
       "      <td>55000.000080</td>\n",
       "      <td>179.751271</td>\n",
       "      <td>0.496394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>15.954841</td>\n",
       "      <td>8.783636</td>\n",
       "      <td>13414.634022</td>\n",
       "      <td>43.957934</td>\n",
       "      <td>0.500288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>32.600000</td>\n",
       "      <td>19.000000</td>\n",
       "      <td>13996.500000</td>\n",
       "      <td>104.780000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>50.615000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>47031.802500</td>\n",
       "      <td>138.550000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>67.940000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>57012.300000</td>\n",
       "      <td>183.480000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>78.555000</td>\n",
       "      <td>42.000000</td>\n",
       "      <td>65470.635000</td>\n",
       "      <td>217.850000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>91.430000</td>\n",
       "      <td>61.000000</td>\n",
       "      <td>79484.800000</td>\n",
       "      <td>269.960000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Daily Time Spent on Site         Age   Area Income  \\\n",
       "count                907.000000  979.000000   1000.000000   \n",
       "mean                  64.794818   36.009193  55000.000080   \n",
       "std                   15.954841    8.783636  13414.634022   \n",
       "min                   32.600000   19.000000  13996.500000   \n",
       "25%                   50.615000   29.000000  47031.802500   \n",
       "50%                   67.940000   35.000000  57012.300000   \n",
       "75%                   78.555000   42.000000  65470.635000   \n",
       "max                   91.430000   61.000000  79484.800000   \n",
       "\n",
       "       Daily Internet Usage  Clicked on Ad  \n",
       "count            913.000000     832.000000  \n",
       "mean             179.751271       0.496394  \n",
       "std               43.957934       0.500288  \n",
       "min              104.780000       0.000000  \n",
       "25%              138.550000       0.000000  \n",
       "50%              183.480000       0.000000  \n",
       "75%              217.850000       1.000000  \n",
       "max              269.960000       1.000000  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "advertising.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Concise Summary of  Data \n",
    "This method prints information about the dataframe such as the index dtype and column dtypes, non-null values and memory usage."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1000 entries, 0 to 999\n",
      "Data columns (total 10 columns):\n",
      " #   Column                    Non-Null Count  Dtype  \n",
      "---  ------                    --------------  -----  \n",
      " 0   Daily Time Spent on Site  907 non-null    float64\n",
      " 1   Age                       979 non-null    float64\n",
      " 2   Area Income               1000 non-null   float64\n",
      " 3   Daily Internet Usage      913 non-null    float64\n",
      " 4   Ad Topic Line             1000 non-null   object \n",
      " 5   City                      1000 non-null   object \n",
      " 6   Gender                    1000 non-null   object \n",
      " 7   Country                   1000 non-null   object \n",
      " 8   Timestamp                 1000 non-null   object \n",
      " 9   Clicked on Ad             832 non-null    float64\n",
      "dtypes: float64(5), object(5)\n",
      "memory usage: 85.9+ KB\n"
     ]
    }
   ],
   "source": [
    "advertising.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Label Encoding\n",
    "\n",
    "In this part _replace_ method is used in oreder to convert categorical data to numerical data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "advertising = advertising.replace({'Gender': {'Male': MALE, 'Female': FEMALE}})"
   ]
  },
  {
   "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>Daily Time Spent on Site</th>\n",
       "      <th>Age</th>\n",
       "      <th>Area Income</th>\n",
       "      <th>Daily Internet Usage</th>\n",
       "      <th>Ad Topic Line</th>\n",
       "      <th>City</th>\n",
       "      <th>Gender</th>\n",
       "      <th>Country</th>\n",
       "      <th>Timestamp</th>\n",
       "      <th>Clicked on Ad</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>68.95</td>\n",
       "      <td>35.0</td>\n",
       "      <td>61833.90</td>\n",
       "      <td>256.09</td>\n",
       "      <td>Cloned 5thgeneration orchestration</td>\n",
       "      <td>Wrightburgh</td>\n",
       "      <td>0</td>\n",
       "      <td>Tunisia</td>\n",
       "      <td>2016-03-27 00:53:11</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>80.23</td>\n",
       "      <td>31.0</td>\n",
       "      <td>68441.85</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Monitored national standardization</td>\n",
       "      <td>West Jodi</td>\n",
       "      <td>1</td>\n",
       "      <td>Nauru</td>\n",
       "      <td>2016-04-04 01:39:02</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>69.47</td>\n",
       "      <td>26.0</td>\n",
       "      <td>59785.94</td>\n",
       "      <td>236.50</td>\n",
       "      <td>Organic bottom-line service-desk</td>\n",
       "      <td>Davidton</td>\n",
       "      <td>0</td>\n",
       "      <td>San Marino</td>\n",
       "      <td>2016-03-13 20:35:42</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>74.15</td>\n",
       "      <td>29.0</td>\n",
       "      <td>54806.18</td>\n",
       "      <td>245.89</td>\n",
       "      <td>Triple-buffered reciprocal time-frame</td>\n",
       "      <td>West Terrifurt</td>\n",
       "      <td>1</td>\n",
       "      <td>Italy</td>\n",
       "      <td>2016-01-10 02:31:19</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>68.37</td>\n",
       "      <td>35.0</td>\n",
       "      <td>73889.99</td>\n",
       "      <td>225.58</td>\n",
       "      <td>Robust logistical utilization</td>\n",
       "      <td>South Manuel</td>\n",
       "      <td>0</td>\n",
       "      <td>Iceland</td>\n",
       "      <td>2016-06-03 03:36:18</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Daily Time Spent on Site   Age  Area Income  Daily Internet Usage  \\\n",
       "0                     68.95  35.0     61833.90                256.09   \n",
       "1                     80.23  31.0     68441.85                   NaN   \n",
       "2                     69.47  26.0     59785.94                236.50   \n",
       "3                     74.15  29.0     54806.18                245.89   \n",
       "4                     68.37  35.0     73889.99                225.58   \n",
       "\n",
       "                           Ad Topic Line            City  Gender     Country  \\\n",
       "0     Cloned 5thgeneration orchestration     Wrightburgh       0     Tunisia   \n",
       "1     Monitored national standardization       West Jodi       1       Nauru   \n",
       "2       Organic bottom-line service-desk        Davidton       0  San Marino   \n",
       "3  Triple-buffered reciprocal time-frame  West Terrifurt       1       Italy   \n",
       "4          Robust logistical utilization    South Manuel       0     Iceland   \n",
       "\n",
       "             Timestamp  Clicked on Ad  \n",
       "0  2016-03-27 00:53:11            0.0  \n",
       "1  2016-04-04 01:39:02            0.0  \n",
       "2  2016-03-13 20:35:42            0.0  \n",
       "3  2016-01-10 02:31:19            0.0  \n",
       "4  2016-06-03 03:36:18            0.0  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "advertising.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Counting Missing Values\n",
    "In this part the combination of two Pandas methods (_isna_ and _sum_) is used in oreder to count number of missing values at each column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Daily Time Spent on Site     93\n",
       "Age                          21\n",
       "Area Income                   0\n",
       "Daily Internet Usage         87\n",
       "Ad Topic Line                 0\n",
       "City                          0\n",
       "Gender                        0\n",
       "Country                       0\n",
       "Timestamp                     0\n",
       "Clicked on Ad               168\n",
       "dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "advertising.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Filling Missing Values\n",
    "Pandas *fillna* method is used to fill out the missing values in the given column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "for column in advertising:\n",
    "    if advertising[column].isnull().values.any() == False or column == TARGET:\n",
    "        continue \n",
    "    advertising[column] = advertising[column].fillna(advertising[column].mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Daily Time Spent on Site      0\n",
       "Age                           0\n",
       "Area Income                   0\n",
       "Daily Internet Usage          0\n",
       "Ad Topic Line                 0\n",
       "City                          0\n",
       "Gender                        0\n",
       "Country                       0\n",
       "Timestamp                     0\n",
       "Clicked on Ad               168\n",
       "dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "advertising.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Another Approach for Handling Missing Values\n",
    "Another way to handle missing values in datasets is to drop the rows having null values.   \n",
    "**Pros:**\n",
    "<ul>\n",
    "    <li>\n",
    "Complete removal of data with missing values results in robust and highly accurate model.\n",
    "    </li>\n",
    "    <li>\n",
    "Deleting a particular row or a column with no specific information is better, since it does not have a high weightage.\n",
    "    </li>\n",
    "</ul>\n",
    "<b>Cons:</b>\n",
    "<ul>\n",
    "    <li>\n",
    "It can cause loss of information and data.\n",
    "    </li>\n",
    "    <li>\n",
    "It works poorly if the percentage of missing values is high (about 30%), compared to the whole dataset.\n",
    "    </li>\n",
    "</ul>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Counting Female & Male Users\n",
    "Pandas *groupby* method is used to split the data into groups based on some criteria."
   ]
  },
  {
   "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>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Gender</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>481</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        count\n",
       "Gender       \n",
       "0         519\n",
       "1         481"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "advertising.groupby(['Gender'])['Gender'].agg(['count'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Counting Users has Clicked on Ad or not"
   ]
  },
  {
   "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>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Clicked on Ad</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.0</th>\n",
       "      <td>419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>413</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               count\n",
       "Clicked on Ad       \n",
       "0.0              419\n",
       "1.0              413"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "advertising.groupby([TARGET])[TARGET].agg(['count'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Counting Female Users Older than Me\n",
    "Pandas *loc* attribute access a group of rows and columns by label(s) or a boolean array in the given dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of Users that are Female and Older than me: 514\n"
     ]
    }
   ],
   "source": [
    "specific_users = len(advertising.loc[(advertising['Age'] > MY_AGE) & (advertising['Gender'] == FEMALE)].index)\n",
    "print('Number of Users that are Female and Older than me:', specific_users)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Using Vectorization\n",
    "In this part Pandas *groupby* buit-in method is used in order make computation vectorized."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Clicked on Ad\n",
      "0.0    31.723326\n",
      "1.0    40.295600\n",
      "Name: Age, dtype: float64\n",
      "vectorization: 2.371550 ms\n"
     ]
    }
   ],
   "source": [
    "tic = time()\n",
    "\n",
    "avg_age = advertising.groupby([TARGET])['Age'].mean()\n",
    "\n",
    "toc = time()\n",
    "\n",
    "print(avg_age)\n",
    "print('vectorization: %f ms' % ((toc - tic)  * 1000))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Using For Loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average Age of Users who did not click on Ad: 31.723325881701896\n",
      "Average Age of Users who clicked on Ad: 40.295599848637366\n",
      "for loop: 244.388580 ms\n"
     ]
    }
   ],
   "source": [
    "clicked_avg_age = 0\n",
    "didnt_click_avg_age = 0\n",
    "\n",
    "clicked_count = 0\n",
    "didnt_click_count = 0\n",
    "\n",
    "tic = time()\n",
    "\n",
    "for index, row in advertising.iterrows():\n",
    "    if row[TARGET] == TRUE:\n",
    "        clicked_avg_age += row['Age']\n",
    "        clicked_count += 1\n",
    "    elif row[TARGET] == FALSE:\n",
    "        didnt_click_avg_age += row['Age']\n",
    "        didnt_click_count += 1\n",
    "        \n",
    "if clicked_count > 0:\n",
    "    clicked_avg_age /= clicked_count\n",
    "    \n",
    "if didnt_click_count > 0:\n",
    "    didnt_click_avg_age /= didnt_click_count\n",
    "    \n",
    "toc = time()\n",
    "\n",
    "print('Average Age of Users who did not click on Ad:', didnt_click_avg_age)\n",
    "print('Average Age of Users who clicked on Ad:', clicked_avg_age)\n",
    "print('for loop: %f ms' % ((toc - tic)  * 1000))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div style=\"text-align: justify;\"> \n",
    "    What is happening in vectorization is that NumPy calls high quality numerical libraries (BLAS for instance) which are very good at vector arithmetic. Therefore, the result of our computation will ready much faster than using for loop. In this examole our computation was about 100 times faster by using vectorization.\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dropping Non-Numeric Columns\n",
    "In this part non-numeric columns are dropped from our dataframe; this is done by using *is_numeric_dtype* method to check for numeric columns and *drop* method to drop the *given* column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "for column in advertising:\n",
    "    if is_numeric_dtype(advertising[column]) == False:\n",
    "        advertising.drop(column, axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plotting Histogram of Characteristics\n",
    "In this part histogram of each column is drawn using Pandas *hist* method. We formatted these histograms using useful methods such as *suptitle*, *xlabel*, and *ylabel* from matplotlib library."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "scrolled": false
   },
   "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"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "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": [
    "colors = ['#b5347e', '#9a34b5', '#1cc32f', '#1691a7', '#de6722', '#e5d700']\n",
    "cursor = 0\n",
    "\n",
    "for column in advertising:\n",
    "    plt.figure() \n",
    "    plt.suptitle('Histogram of ' + column, fontsize = 15)\n",
    "    plt.xlabel(column, fontsize = 12)\n",
    "    plt.ylabel('Frequency', fontsize = 12)\n",
    "    advertising[column].hist(color = colors[cursor])\n",
    "    cursor += 1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dropping *Gender* Column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "advertising.drop('Gender', axis=1, inplace=True)"
   ]
  },
  {
   "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>Daily Time Spent on Site</th>\n",
       "      <th>Age</th>\n",
       "      <th>Area Income</th>\n",
       "      <th>Daily Internet Usage</th>\n",
       "      <th>Clicked on Ad</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>68.95</td>\n",
       "      <td>35.0</td>\n",
       "      <td>61833.90</td>\n",
       "      <td>256.090000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>80.23</td>\n",
       "      <td>31.0</td>\n",
       "      <td>68441.85</td>\n",
       "      <td>179.751271</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>69.47</td>\n",
       "      <td>26.0</td>\n",
       "      <td>59785.94</td>\n",
       "      <td>236.500000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>74.15</td>\n",
       "      <td>29.0</td>\n",
       "      <td>54806.18</td>\n",
       "      <td>245.890000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>68.37</td>\n",
       "      <td>35.0</td>\n",
       "      <td>73889.99</td>\n",
       "      <td>225.580000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Daily Time Spent on Site   Age  Area Income  Daily Internet Usage  \\\n",
       "0                     68.95  35.0     61833.90            256.090000   \n",
       "1                     80.23  31.0     68441.85            179.751271   \n",
       "2                     69.47  26.0     59785.94            236.500000   \n",
       "3                     74.15  29.0     54806.18            245.890000   \n",
       "4                     68.37  35.0     73889.99            225.580000   \n",
       "\n",
       "   Clicked on Ad  \n",
       "0            0.0  \n",
       "1            0.0  \n",
       "2            0.0  \n",
       "3            0.0  \n",
       "4            0.0  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "advertising.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Normalizing Data\n",
    "In this part values of each column is normalized; this is done by subtracting each column value by the mean and then dividing the result by the standard deviation of that column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "for column in advertising:\n",
    "    if column != TARGET:\n",
    "        advertising[column] -= advertising[column].mean()\n",
    "        advertising[column] /= advertising[column].std()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>Daily Time Spent on Site</th>\n",
       "      <th>Age</th>\n",
       "      <th>Area Income</th>\n",
       "      <th>Daily Internet Usage</th>\n",
       "      <th>Clicked on Ad</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.273474</td>\n",
       "      <td>-0.116122</td>\n",
       "      <td>0.509436</td>\n",
       "      <td>1.817578</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.015870</td>\n",
       "      <td>-0.576377</td>\n",
       "      <td>1.002029</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.307698</td>\n",
       "      <td>-1.151696</td>\n",
       "      <td>0.356770</td>\n",
       "      <td>1.351152</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.615713</td>\n",
       "      <td>-0.806505</td>\n",
       "      <td>-0.014448</td>\n",
       "      <td>1.574722</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.235301</td>\n",
       "      <td>-0.116122</td>\n",
       "      <td>1.408163</td>\n",
       "      <td>1.091154</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Daily Time Spent on Site       Age  Area Income  Daily Internet Usage  \\\n",
       "0                  0.273474 -0.116122     0.509436              1.817578   \n",
       "1                  1.015870 -0.576377     1.002029              0.000000   \n",
       "2                  0.307698 -1.151696     0.356770              1.351152   \n",
       "3                  0.615713 -0.806505    -0.014448              1.574722   \n",
       "4                  0.235301 -0.116122     1.408163              1.091154   \n",
       "\n",
       "   Clicked on Ad  \n",
       "0            0.0  \n",
       "1            0.0  \n",
       "2            0.0  \n",
       "3            0.0  \n",
       "4            0.0  "
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "advertising.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>Daily Time Spent on Site</th>\n",
       "      <th>Age</th>\n",
       "      <th>Area Income</th>\n",
       "      <th>Daily Internet Usage</th>\n",
       "      <th>Clicked on Ad</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.000000e+03</td>\n",
       "      <td>1.000000e+03</td>\n",
       "      <td>1.000000e+03</td>\n",
       "      <td>1.000000e+03</td>\n",
       "      <td>832.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.101341e-16</td>\n",
       "      <td>2.184919e-16</td>\n",
       "      <td>-3.854694e-16</td>\n",
       "      <td>-1.740830e-16</td>\n",
       "      <td>0.496394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>0.500288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-2.118908e+00</td>\n",
       "      <td>-1.957144e+00</td>\n",
       "      <td>-3.056625e+00</td>\n",
       "      <td>-1.785019e+00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-7.891122e-01</td>\n",
       "      <td>-8.065049e-01</td>\n",
       "      <td>-5.939929e-01</td>\n",
       "      <td>-9.032985e-01</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>5.430951e-02</td>\n",
       "      <td>-1.161217e-01</td>\n",
       "      <td>1.500078e-01</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>8.432692e-01</td>\n",
       "      <td>6.893254e-01</td>\n",
       "      <td>7.805382e-01</td>\n",
       "      <td>8.247266e-01</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.753000e+00</td>\n",
       "      <td>2.875539e+00</td>\n",
       "      <td>1.825231e+00</td>\n",
       "      <td>2.147814e+00</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Daily Time Spent on Site           Age   Area Income  \\\n",
       "count              1.000000e+03  1.000000e+03  1.000000e+03   \n",
       "mean               1.101341e-16  2.184919e-16 -3.854694e-16   \n",
       "std                1.000000e+00  1.000000e+00  1.000000e+00   \n",
       "min               -2.118908e+00 -1.957144e+00 -3.056625e+00   \n",
       "25%               -7.891122e-01 -8.065049e-01 -5.939929e-01   \n",
       "50%                5.430951e-02 -1.161217e-01  1.500078e-01   \n",
       "75%                8.432692e-01  6.893254e-01  7.805382e-01   \n",
       "max                1.753000e+00  2.875539e+00  1.825231e+00   \n",
       "\n",
       "       Daily Internet Usage  Clicked on Ad  \n",
       "count          1.000000e+03     832.000000  \n",
       "mean          -1.740830e-16       0.496394  \n",
       "std            1.000000e+00       0.500288  \n",
       "min           -1.785019e+00       0.000000  \n",
       "25%           -9.032985e-01       0.000000  \n",
       "50%            0.000000e+00       0.000000  \n",
       "75%            8.247266e-01       1.000000  \n",
       "max            2.147814e+00       1.000000  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "advertising.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plotting PMF of Characteristics \n",
    "In this part the probability density function is drawn for each characteristic. In each plot, We drew two curves in each figure. These curves indicate whether the user has clicked on ad or not."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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": [
      "Clicked std 0.8361610114854059\n",
      "not clicked std 0.5313689387678912\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAEmCAYAAACNq4wIAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/d3fzzAAAACXBIWXMAAAsTAAALEwEAmpwYAAAw80lEQVR4nO3de7xUdb3/8deHDbJBUFPxsgUEL5kgeNtgmqaBmqhhZiJqCqVQRynreOxklpJlaXLKMszwmooaihpHJRUUEQvbmw5KoPhD4y4ISChukNvn98dae5g9ey5r9p7Lnpn38/GYB7O+a82az9qsWZ+1vt+1vl9zd0RERADaFTsAERFpO5QUREQkRklBRERilBRERCRGSUFERGKUFEREJEZJoZXMbKyZedxrpZlNNrODc7T+GWb2eI7WtdjMxmVY5pRwO46IK3MzGxM3PdrMvpyjmHol/P0+NrN3zGyimZ2Ui+9oRWyF3O74V/dcfEcWsewT7se9Esqb7QttgZkdbmavhPuKJ8adZPnG7ZhVoBBLWvtiB1AmNgBnhO8PAn4KTDezvu7+cfHCapF/AMcD76RZZjTwT+CpHH7vfwGvAh2B3sBwYKaZjXX3n+Twe7JxPPCvuOl8bne893O4/ij2AW4AZgCL48qj7AvFcCuwBzAU+Bh4L8PyF4b/nmBmPd19aR5jK3lKCrmxzd1nh+9nm9lS4BXgTOCxxIXNrJO7bypkgFG5+4fA7IwL5t7CuL/hy8D9ZnYjMNbMXnb3GYUOKC6efFpYoO/JWhH3hUw+A0xx9+mZFjSzDsBXgReBQQQnG7/Mb3ilTdVH+TEn/LcXxKpt/sfMfmxmy4EPw/LOZvZbM1tlZpvNrM7MTk+2wrDqYrGZbTKzZ8zsgIT5N5vZPDPbaGbLw+qX/VKs68fhd24Ml9s9bl7aKgMzmwEcC4yIq+4YaWa/NLN3zcwSlh9pZlvMrFuUP1yCnwArgW8lrPMcM6sP/2arwu/uEDd/rJmtNbOjzWy2mTWY2f8lVkeZ2VAzmxNWQ6w3s9fM7OS4+bHqowJvN2Z2v5nVJ5Q1VjmdnRDjVWb2czNbY2bvm9l4M+uY8NkDzeyR8O/SYGZvmNlFYdXLvHCxlxq3LfxMsqrEjPushVWe4foXmdmHZjbVIlSLmdlRZjY9jHF9uH/uG7/9wMHA98LYZmRY5enAnsAtwN/YedUQ/51mZj8N/3Yfmtm9ZjbcEqqmzKw6/P9eZmafmNnrZnZmpm0qOe6uVytewFhgbULZ4YADl4TTiwkucacRXPJ+JSyfCHwEfBsYAjwBbAVOjFvXDGAFwQ/3K8BFwDKgLuE77yXY4U8mODP6G7AAaBe3zOJwXS8DXyKoDvk38FjcMqeEsR8RV+bAmPB9H+BN4Bngs+GrG8HZmwOnJMT1MjA5zd+vV/i5s1PMfxBYHDc9DNgO3EHwg/+PcBvGJfyfNABvAF8P/7azgTVA53CZg4EtBFURgwiu6n4MnFvg7R5KcMXe+GoXzr8fqM/0twqnl4bLfxG4BtgGfD9umX0IkusiYCQwGLgK+G+C6rqLwvVc0bhtafaFqPvsMuCvwDkEZ+ergWcz/Ja6hf+XfwO+DHwNWB7+P+4SxvpZgt/SxPB9nwzrfIigOq4qjNmBzyQs8z2CfepGgn1qfPg3daBX3HJPh+v6j3C5u8O/9VHFPg7l8lX0AEr9RZgU4n7UnwZeIrga2D9cZnG4I1fHfe5wYAcwIq6sHUGd9XNxZTPCH13PuLLPhTvsGSliqgIOCJf5fFz5YuADoEtc2cVhHIeH08kOBLGDYzhdD9yf5HtnAX+Mmz4oXHfSA364TC/SJ4VfAJvC9wYsAe5LWOYbwCZgr7j/EwcGxS1zVPzfjCBxrsvwf1uI7U58PRTOv5/oSWFmwnJPAbMT/oYfN+6PSWI5guSJrcm+QHb77AbgU3Fl3w3X1SnN3+RmgqSwW1zZceHnLkzYj8elWk/ccp0IEtgd4fS+BAfxnyT8Vt4Dxid89lnikgJBInXg5ITlZhJ3UlUOL1Uf5cZeBAfurcBCgoPCBe4e3wA23d03x00PIDjIxdoc3H1HOH1iwvr/4XGNY+7+KsEZy8DGMjMbYmZ/NbMNBDv+8nDWpxPW9YK7b4ybfjKMY0DEbU3nHuA8M+sSTo8kOEP8SyvWGV8t82mgJzDJzNo3vgjqi6sJDm6NthAcnBotCP9trMKYB+xuZn80s9PNbNdWxNia7f4ewd++8fXjFnz/8wnTC9i5nRBcCf0lYX9siWz22Tp3X58QEwQnK6kMBJ73oC2jcf2vESSBxPVH8SWgC/BouK7VBPtEfBVSD2A/YErCZxOnTwVWAa8m7HvTgdoWxNZmKSnkxgaCH0wtwY+xl7tPTVhmdcL0/sBGd29IslznhDrhZHejvB+uAzMbQLATLwcuIbhj5LPhctVJPhcTfv/GxnW10iSCM8lhYR37COABd9/WinUewM6/3d7hv8+yMwlvZecdQj3iPvdReMACwN23hG+rw+mFBFUbB4XrW2tmD7ewDaA1273I3evjXv/K/JFm/p0wvYWm/+97kfkOnSiy2WeTxQTN98fE9Sf+ThrXv2cWcTa6MPzsPDPbw8z2AP4XONTMjg2XaWx3W5Pw2cTpvcNltya8xtJ0vyt5uvsoN7a5e32GZTxh+j2gi5l1TviR7Qs0uPsncWX7JFnfPuz8oZ9LsBNf4OE1rZkdmCKOJusys84EZ1OtPmi4+8dm9ijBmfISgrP6+1q6vvBMbBDBJToEVV8QtIX8X5KPZHVAdfdngGcsaGg/C7gNuJ2gDjyb9eR0u0ObCerR432qhetaR26Sfjb7bEvXn2xf35edN29EEv6fDiFoh/ggySIXhutcFU4nngwkTn9A0B735WziKEW6UiieOoJE8dXGgvAs86sEddTxjjGznnHLfY7gx/P3sKgTsLUxIYQuTvG9p8VVc0CQUJygvjyqxDPRePcAJxGcQc1297eyWG+i64Ea4M5weiHBD7NXwtl142tdS77E3Te4+8MEVWl90ixaqO2G4Kqvl5nFf1/SO9MimA58sfEuniSinMVDdvtsS7xGEGfXuPUPIGhLyXb9XyFICCOALyS8ngcuCGNfRpAYzkn4/NCE6ekEVwobk+17WcbWpulKoUjc/U0zewT4XfgjeAcYRXA3y38kLL6G4Iz2BoIf7i0E7QyNddYvAN81s9sILo9PILhzI5lN4bpuJTh7vBV40t0XpFg+mbcIfrxfJDgL/VfjAdndXzOz+QR1wN/MYp2HmdlagrPjxofXzgDGuvvL4bp3mNnVwINmthswleCAdhDBGdxXk1RtJGVm3ySoZvsLwZ05hwLnAw8UeLtTeYrgbpi7zex+4GiCBvWW+DVwKfCKmd1EcCA8HNjV3X9JcKfNJoLbbTcQnGA0O9Bluc+2xK/C9TxnZrcQXMHeTND+MznLdV0IvOXuzf4/zWzPcH0nufvM8Ldwq5mtIXiQcCjQL1y8sQryBeA54IUwtvnAbgQ3MFS7+7VZxtd2Fbulu9RfJLklNckyi0lytwTQmaC6YjXwCcHZ+hcTlpkBPE5wr37jj3cq0CNhue8T/Ng/Jrj19VCa3z2zGPifMObV4bKPAHvELXMKme8+Oij8jg3hvJEJsfyM4JbQ3dL9XcJle9H07ptNwLsEtxyelOIzQwgeDvyY4C6vueF3tk/3fxK/HQQJ4RmChLCZoOrpFqBjgbc73R1KIwkOvA0Et0OekPiZxBhTbT9wIPAnYH24vteB4XHzLwbeJkiynmZfiLzPJpQ1W1eKbT6a4MaBBoJ2iYeBfaP8nuLmN95l9MMU8zuGf4ffh9MW/t+tIbhbaSJBcnKa/jY6Ejw7syj8O60iOKk4q7XHkbb0snBjRXLGzP5O8KTuJcWOpZAqdbvLkZndDZzm7qna5sqWqo8kZ8yslqBheABwZZHDKZhK3e5yET6xfQHBw3Y7CK5Ev07wcF/FUVKQXKojuOS/1t3rihxLIVXqdpeLjwnagsYAuxLcQfbfBFWtFUfVRyIiEqNbUkVEJEZJQUREYpQUREQkRklBRERilBRERCRGSUFERGKUFEREJEZJQUREYpQUREQkRklBRERilBRERCRGSUFERGJKupfUvffe23v16lXsMERESsqcOXPWunviONRAiSeFXr16UV9fVsOjiojknZktSTVP1UciIhKjpCAiIjFKCiIiElPSbQoiUlq2bt3K8uXL2bx5c7FDqQjV1dV0796dDh06RP6MkoKIFMzy5cvp2rUrvXr1wsyKHU5Zc3fWrVvH8uXL6d27d+TPKSlIwdw14NlmZaPqzixCJPmz0iY1K6vxYUWIpG3avHmzEkKBmBl77bUXa9asyepzSgqSM+kO+snmNZaXS2JIlhAayxsTg5IGSggF1JK/tRqaJSfSHfQlkC5piLQVSgoiUnFWrVrF8OHDOfjggzn22GM588wzefvttzniiCMAqK+v5zvf+U7adXTp0qXF39+rVy/Wrl3b4s/nk6qPpE0opfYGVQGVNnfn3HPPZcSIETz66KMAvP7666xevTq2TG1tLbW1tcUKsah0pSBFV0pVT6oCKqyGiUtY3etpVrabxOpeT9MwMWXvDJG99NJLdOjQgW9961uxsiOPPJIePXrEpmfMmMHZZ58NwMaNG/n6179Ov3796N+/P5MnT26yvrVr13L88cfzzDPPsGbNGs477zwGDBjAgAEDePXVVwFYt24dp59+On379uXyyy/H3Vu9HfmipCAFkeqsv61eDbREqqsFXUW0TMPEJWwYXc/2JQ3gsH1JAxtG17c6Mfzzn//k2GOPjbz8T3/6U3bffXfmzZvHG2+8waBBg2LzVq9ezVlnncWNN97IWWedxVVXXcX3vvc96urqmDx5MpdffjkAP/nJTzjxxBOZP38+5557LkuXLm3VNuSTqo8kJ0bVnZmxCqicEkAq6RJAjQ9T1VMWPrpuHt6wvUmZN2zno+vm0fniAwsWx7Rp02LVTACf+tSngOBBvMGDBzN+/HhOPvnk2LILFiyILfvhhx+yceNGZs6cyRNPPAHAWWedFVtHW6SkIFlJd+CvhIN+a6VLAEoYTW1f2pBVeVR9+/bl8ccfb9U6ANq3b8+xxx7Lc889F0sKO3bsYPbs2VRXV7d6/cWi6iOJrJTq/kuN2iqaq+rZOavyqAYNGsQnn3zChAkTYmVvvPEGy5YtS7r8aaedxvjx42PT69evB4JnAO69917eeustbrnlFgBOP/10br/99tiyc+fOBeDzn/88Dz/8MABTp06NraMtUlKQoiul9ga1GxRO15v6YZ2rmpRZ5yq63tSvVes1M5588kmmTZvGwQcfTN++fbn22mvZb7/9ki7/ox/9iPXr13PEEUdw5JFH8tJLL8XmVVVV8cgjj/Diiy9yxx138Nvf/pb6+nr69+9Pnz59uPPOOwG44YYbmDlzJn379uWJJ56gZ8+erdqGfLK23AqeSW1trWuQncJJd0XQFg/gpSTdFUE5JZw333yTww8/PPLyDROX8NF189i+tIGqnp3pelO/grYnlINkf3Mzm+PuSe+5VZuCiLRZnS8+UEmgwJQUpM0rxoNtavSVSqU2BYmsGHX/xWjcLkajr9oqpK3QlYJkRW0H+aMEIG2BrhRERCRGSUFERGKUFERKwEqb1OwlLWNmXH311bHpcePGMXbs2LSfeeqpp5p0X5HogQce4IgjjqBfv34cffTRjBs3DoCRI0fGnp6+/PLL065j7Nixsc9l6/7772fMmDEt+mwiJQVp5q4BzzZ7FUsxGrfbWqOvnnbOrY4dO/LEE09kNZ5BuqQwdepUbrvtNp5//nnmzZvH7Nmz2X333Zstd/fdd9OnT58Wx10oSgrSRFvsymJU3ZnNXvlW48OavaTwFk1dwSNfepG7Bj7LI196kUVTV7R6ne3bt2f06NH8+te/bjZv8eLFDBo0iP79+zN48GCWLl3KX//6V6ZMmcI111zDUUcdxTvvvNPkM7/4xS8YN24cNTU1QJB0Ro0a1Wzdp5xyCo0P2/7lL3/hmGOO4cgjj2Tw4MHNlr3rrrsYMmQImzZt4qGHHmLgwIEcddRRfPOb32T79qCTwPvuu49Pf/rTDBw4MNZFdy4oKYhIm7Ro6gpe+fk8Nq7aDA4bV23mlZ/Py0liuPLKK5k4cSIbNmxoUv7tb3+bESNG8MYbb3DxxRfzne98hxNOOIGhQ4dy6623MnfuXA4++OAmn8m2K+41a9YwatQoJk+ezOuvv85jjz3WZP7vfvc7nn76aZ566ikWL17Mn/70J1599VXmzp1LVVUVEydO5L333uOGG27g1VdfZdasWWmrpbJVsKRgZmeY2UIzW2RmP0gyf6SZrTGzueHr8kLFJiJtT90dC9m2eUeTsm2bd1B3x8JWr3u33Xbj0ksv5be//W2T8r/97W9cdNFFAFxyySXMmjWr1d+VaPbs2Xz+85+nd+/eAOy5556xeQ888ABTp07l8ccfp2PHjkyfPp05c+YwYMAAjjrqKKZPn867777La6+9ximnnEK3bt3YZZdduOCCC3IWX0GSgplVAeOBIUAf4EIzS1a59id3Pyp83V2I2ESkbdq4enNW5dn67ne/yz333MPHH3/cqvX07duXOXPm5CSmfv36sXjxYpYvXw4EQ4eOGDGCuXPnMnfuXBYuXJixUby1CnWlMBBY5O7vuvsW4FHgnAJ9t0hJa2sN34XSZd/kYxKkKs/WnnvuybBhw7jnnntiZSeccEJsQJ2JEydy0kknAdC1a1c++uijpOu59tprueaaa1i1ahUAW7Zs4e67U5/Tfvazn2XmzJn861//AuCDDz6IzTv66KP5wx/+wNChQ1m5ciWDBw/m8ccf5/33348tu2TJEo477jhefvll1q1bx9atW5tVQbVGoZLCAUB8Z+XLw7JE55nZG2b2uJn1SDJf8qyUurGuJJXY8D3gisNoX930ENW+uh0DrjgsZ99x9dVXN7kL6fbbb+e+++6jf//+PPjgg/zmN78BYPjw4dx6660cffTRzRqazzzzTMaMGcOpp55K3759OeaYY/jwww9Tfme3bt2YMGECX/nKVzjyyCObVf2ceOKJjBs3jrPOOot99tmHn/3sZ5x++un079+f0047jffee4/999+fsWPHcvzxx/O5z30uq55nMylI19lm9lXgDHe/PJy+BDjO3cfELbMXsNHdPzGzbwIXuPugJOsaDYwG6Nmz57FLlrR+IG8pXcXoLE9aLtuusxdNXUHdHQvZuHozXfatZsAVh3HIkGTnk5JKW+06ewUQf+bfPSyLcfd1cZN3A79MtiJ3nwBMgGA8hdyGKaUk3e2zURKDekJt+w4ZcoCSQIEVqvqoDjjUzHqb2S7AcGBK/AJmtn/c5FDgzQLFJhVID4SJJFeQKwV332ZmY4DngCrgXnefb2Y3AvXuPgX4jpkNBbYBHwAjCxGbiBSWu2NmxQ6jIrSkeaBgXWe7+7PAswll18e9vxa4tlDxiEjhVVdXs27dOvbaay8lhjxzd9atW0d1dXZ3a2k8BREpmO7du7N8+XLWrFlT7FAqQnV1Nd27d8/qM0oKUrJG1Z2pu49KTIcOHWJP8krbpKRQocrlYNrSmGt8WNncfVQu2yFtQ0GeU8iX2tpab+x1UKJL1+NpKSaGSpbubiklBkkl3XMK6iVVRERilBRERCRGSUFERGKUFEREJEZJoQKpJ9TyUandakv+6JbUCqUEUD6UACSXdKUgIiIxSgoiIhKj6iMpW3cd+zQkdrq2wxn1j7OLE5BICdCVgpSlWEJIfLUz7jrm6WKHJ9Jm6UpBylNjEkhWrlMhkZT08xARkRglBRERiVFSkPLkHrySle8o3Z6BRfItUlIws6vMbO98ByOSK6PmnL0zMcS/dPeRSFpRG5oHATeZ2QzgQeApd/8kb1GJ5MCoOTr4i2QrUlJw93PMbC9gOPBd4E4zmww84O4z8xiftFC5jKwmraeR2SQbkdsU3H2du4939+OBk4EBwEtmttjMrjOzLnmLUrKSamS1dCOuSXlKNTJbuhHbpLJl1dBsZoPN7D5gBrAauBS4BDgamJrz6EREpKAiVR+Z2TiCqqMNwAPAj9x9Rdz82cD6vEQoIiIFE7WhuRo4193rks10961mlnQQaBERKR1Rq492JEsIZnZb43t3fytXQYmISHFETQojU5RfkqM4JIc0spo00shskq201Udm9o3G5eLeNzoIWBv1i8zsDOA3QBVwt7vfnGK584DHgQHuXh91/dKUEoA0UgKQbGRqU2i8EtiFplcFTnD30YgoX2JmVcB44DRgOVBnZlPcfUHCcl2Bq4DXoqxXRERyK21ScPcvAJjZz9z9R634noHAInd/N1zfo8A5wIKE5X4K3AJc04rvEhGRFkqZFMzM3GM9il1vZknbH9x9R4TvOQBYFje9HDgu4fuOAXq4+zNmljIpmNloYDRAz549I3y1SHJ66lukuXRXChuA3cL32wiqjOJZWFbV2iDChPMrUjdox7j7BGACQG1trbq7rHAt7cIh3VPfSgxSydIlhb5x73u38ntWAD3ipruHZY26AkcAMywYLWs/YIqZDVVjs6SSrgsHNa6KtEzKpODuy+LeL4mfZ2adCJ5diNpTah1wqJn1JkgGw4GL4ta/AYh1zR32xvpfSggiIoUVdTyFcWY2MHx/FvABsN7MvhTl8+6+DRgDPAe8CUxy9/lmdqOZDW1Z6CIikmtRu7m4GLg+fH898DWCNodfA/8bZQXu/izwbELZ9SmWPSViXCIikkNRn2ju7O4N4ZgKB7n7ZHefBhyYx9hE8kZPfYskF/VK4W0zuxg4BHgBIByec1O+AhPJpMaHtWoAGSUAkeaiJoUrCLqo2AJcFpZ9EXg+H0GJRKW7jERyK+pwnHXACQllE4GJ+QhKotHDV9JaKzs/BpviHvfpZNQ0nF+8gKToIo+8ZmaHmdkwM/tG/CufwUlqGnJTWqtZQgDY5EG5VKyoI6/9kOCuo9eBhrhZDtybh7hEJN8SE0KmcqkIUdsUvgsMdPc38hiLiIgUWdTqo02ARlYTESlzUZPCj4HbzWx/M2sX/8pncCKSR50su3KpCFEP6vcDowi6vN4avraF/0oR6OEraa2ahvObJwDdfVTxorYptLaXVMkDJQBpLSUASRT1OYUlEBv3YF93fy+vUYmISFFE7SV1DzN7GNgMLArLhprZz/IZnIiIFFbUNoU7CXpFPZCgqwuAvwEX5CMoEREpjqhtCoOBGnffamYO4O5rzGyf/IUmEmhNp3ctpS5EpFJFvVJoMjIagJn1BNS2IHmVbsjNfFEXIlLJoiaFu4HJZvYFoJ2ZHQ/8kaBaSUREykTU6qNbCJ5qHg90IOjv6A8E3WmLiEiZiHpLqhMkACUBEZEyFrWX1DOBk4A9gQ+Ame4+NZ+BiYhI4aVtUzCzXcxsGvAYwSA7uwGfAx43s+lmtksBYpQKluouo3zefaQuRKSSZbpS+B6wF/AZd1/WWGhmPYCngP8Ebs5bdCIUZ8hNJQCpVJmSwleBq+ITAoC7LzOz/wR+hZKCSFkqxvMhUnyZksKhQF2KeX8HDsltOJJID1FJMaR7PkSJobxlek7B3H1TshmpyiV39BCViBRapiuFXczs60CqUTc65DgeEREpokxJ4TXg0gzzIzGzMwiec6gC7nb3mxPmfwu4EtgObARGu/uCqOsXEZHWS5sU3P2UXHyJmVURPA19GsHobXVmNiXhoP+wu98ZLj+UoBH7jFx8v4iIRFOoMZYHAovc/V133wI8CpwTv4C7fxg3uSvgBYpNRBIU4/kQaRui9n3UWgcA8be1LgeOS1zIzK4kePZhF2BQshWZ2WhgNEDPnj1zHmhbMqruTN19JEWjBFCZCpUUInH38cB4M7sI+BEwIskyE4AJALW1tWV/NaEEICKFVKjqoxVAj7jp7mFZKo8CX85nQCIi0lzUMZqfNLMvm1lLb0GtAw41s95hf0nDgSkJ33Fo3ORZwP9r4XeJiEgLRa0+egW4HrjHzCYBD7r7X6N+ibtvM7MxwHMEt6Te6+7zzexGoN7dpwBjzOxUYCuwniRVRyJtgdp5pJxZMFRCxIXN+gJfAy4CtgAPAhPd/Z38hJdebW2t19fXF+OrJcdKpZ+ddE+TKzFIqTCzOe5em2xeVm0K7j7f3a8lSAwNwA3AP8xsmpkd2fpQpRIVYxxmEUkuclIws8PM7Kdm9g7B3T9/AnoB+wLPEnSlLSIiJSzqyGv1BAngT8BF7p7YvcWvzOzbOY5NREQKLGpD883AlPBp5KTcvXduQhIRkWKJWn10XbKEEF5BiFQMDdUp5S7qlcLBiQVmZsBBuQ1HKlGNDyuZu49ACUDKW9qkYGYPhG87xr1v1AuYn4+gpPK01QQgyZVSEpfsZLpSeCfFewdeBR7LeUQVSA9DSSnRUJ3lLdN4Cj8BMLPZ7v5cYUKqLOmG3FRiEJFCS5kUzCy+6+qtCdMx7v5izqMSEZGiSHelcE+EzztqbBYRKRspk4KeOxARqTyFGk9BRMqEhuosb+naFN5098PD98tIMWayu5f3mJh5piE3pRQpAZSvdG0Ko+Lefy3fgVQyJQARaSvStSnMinv/cmHCERGRYoraS+ouwI+AC4EaYCXBOMo3ufvm/IUnUlpUFSilLmrfR78HDgO+AywBDgR+CBwAfCM/oUm5WVk1CXbEFbSDmu3lUzetBxGlHERNCl8GDnb3f4fTC8zsNWARSgoSQbOEALAjKC+nxCBS6qLekroK6JxQ1gl4L7fhSNlKTAiZykWkKKJ2c/Eg8Bczux1YDvQArgQSe04VEZESlm03Fz9MmP4mcEvuwhERkWJSNxdSGO1IXlVURs/U60FEKQdRG5pFWqVm+7Cyv/sIlACk9EV9TmE3YCxwMrA3YI3z1M1FZjp7DJRbApDUNDJb6Yp68X4HcAxwI7An8G1gKfDrPMVVNtLduy5SjtKNzCZtX9SkcDpwnrv/Gdge/nsBcEnULzKzM8xsoZktMrMfJJn/n2a2wMzeMLPpZnZg1HWLiEhuRE0K7YAN4fuNZrY7wTMKh0T5sJlVAeOBIUAf4EIz65Ow2P8Bte7eH3gc+GXE2EREJEeiJoXXCdoTAF4hqE76PfB2xM8PBBa5+7vuvoWg36Rz4hdw95fcvSGcnA10j7huERHJkahJYRSwOHx/FbAZ2AO4NOLnDwCWxU0vD8tSuQyYmmyGmY02s3ozq1+zZk3ErxcRkSgiJYXwDP+d8P377n6Zu1/g7gtyHZCZfQ2oBW5NEcsEd69199pu3brl+utzLtVdRpV495FUBo3MVtoiP6dgZt+gedfZ97p70hHZEqwg6BqjUfewLPE7TgWuA05290+ixtbWKQFIpVECKF1Rn1P4JUEbwG3s7Dr7vwi60/5+hFXUAYeaWW+CZDAcuCjhO44G/gCc4e7vR4xfRERyKOqVwkjgGHdf3lhgZk8D/yBCUnD3bWY2BngOqCK4wphvZjcC9e4+haC6qAvwmJkBLHX3odlsjBRXJTyx3Fr3nDiVHZ/svLhu19G4bNaQIkYk0pRFqf0xs3cIksKGuLI9gDnufnD+wkuvtrbW6+vri/X1EifpeAmgxBAnMSE0UmKQQjOzOe5em2xeuq6zD4qbvA14wsxuZmfX2degJ5qlkcZLyChZQkhXLlIM6aqPFgFOXD9HwBcSlhkE/C7XQYmISHGk6zq7jDo1FhGRKLI68JtZTzM73sx6ZF5aKkqqPUmnFjHtOlpW5SLFEOkna2b7m9nLBFVKTwDvmNlMM6vJa3RSMmq2D2u+N6mRuYnLZg1plgDUyCxtTdRbUn9P0P/Rme7+sZntCvwcuBPQbaNozARQAohCCUBjLbR1US/uTwSudvePAcJ/vw+ckK/ASonGTBCJRmMttH1Rk8J6gi6v4x0G/Dun0YiISFFFrT76JTDNzO5hZzcXXwd+nK/ARESk8CIlBXe/K3yq+SKgP0GHeBe5+/R8BiciIoWVMSmEo6a9DfRx9xfzH5KIiBRLxjYFd98ObAeq8x9OadKYCSLRaKyFti9qm8JtwCQz+zlB30exzlrc/d08xFVylABEolECaNuiJoXG/o1OSyh3gq6wpULoHvP8eWjINDat3RKb7rT3Lnxt6qlFjEgqUdThONuleCkhVBDdY54/iQkBYNPaLTw0ZFqRIpJKlTYpmFlnM/u5mU0xs7Fm1rFQgYlUksSEkKlcJF8yXSmMB74EvAV8FRiX94hERKRoMiWFM4DT3f37wBDg7PyHJCIixZKpoXlXd38PwN2XmdnuBYipzbr/C8+xdeP22HSHLlWMfOmLRYxIykWnvXdJWlXUae9dihBN8azs/BhsihuJrpNR03B+8QKqQJmuFNqb2RfMbJCZDUqcDssqQmJCANi6cTv3f+G5IkVUeLrHPH++NvXUZgmg0u4+apYQADZ5UC4Fk+lK4X3g3rjpdQnTDhxEBUhMCJnKy5USQP5UUgJIKjEhZCqXvEibFNy9V4HiEBGRNkCDJYqISIySQkQduiR/Ti9VuYhkqVOKsapTlUteKClENPKlLzZLALr7SCR3ahrOb54AdPdRwUXt+0igYhKA+jdqex4b9jL//tfHsek9eu/K+ZNOLmJE+aEEUHwFu1IwszPMbKGZLTKzHySZ/3kz+4eZbTOzrxYqLmlK/Ru1PYkJAeDf//qYx4a9XKSIpJwVJCmEA/WMJ3gqug9woZkljvm8FBgJPFyImERKRWJCyFQu0hqFqj4aCCxqHHvBzB4FzgEWNC7g7ovDeTsKFJOIiCQoVFI4AFgWN70cOK4lKzKz0cBogJ49e7Y+sgSVUncrUmrU1lUYJXf3kbtPcPdad6/t1q1bTtetultpi/bovWtW5eVIbV2FU6iksALoETfdPSxrU1R3q/6N2qLzJ53cLAHoClbypVDVR3XAoWbWmyAZDAcuKtB3S5aUANoeJQAplIJcKbj7NmAM8BzwJjDJ3eeb2Y1mNhTAzAaY2XLgfOAPZja/ELGJiMhOBXt4zd2fBZ5NKLs+7n0dQbVS0ezRe9ekVUWVVHcrIpVNTzTHOX/SyRVz95Hu5CgflbDP1vgw7bMFYu6l21d5bW2t19fXFzuMkpPujg39yEpLsjvmoDwTg+SOmc1x99pk80rullQR2Ul3zEmuVVz10UNDpjUZC7fShjwUKVcrqyZBfH8I7aBmu658s1VRVwqJCQFg09otPDRkWpEiEpFcaJYQAHaE5ZKVikoKiQkhU7lIW6ennUOpekxTT2pZq6ikIAE9tVw+9LSz5FrFtSlUipWdH4NNcXeWJYxgpQRQPjIlgEVTV1B3x0I2rt5Ml32rGXDFYRwy5IACRSelpqKuFDrtvUtW5aWqWUIA2ORBuVSURVNX8MrP57Fx1WZw2LhqM6/8fB6Lpra5rsdaJ9WRrKKOcLlRUX+yr009tVkCKMu7jxITQqZyKVt1dyxk2+amFevbNu+g7o6FRYooP2q2D2t+NNPdRy1ScdVHZZcARNLYuHpzVuWlLF0CyFSdKjtV1JWCSKXpsm91VuXlSNWp2VFSKEedLLtyKVsDrjiM9tVNf+btq9sx4IrDihRREag6NSsVV31UTlJdEtc0nK/LZQGI3WWU6u6je06cyo5Pdu4n7Toal80aUpRYpW1Qh3glKuklMejgL5ElJoRG5ZYY1AFkc+k6xNOVQqnSJbG0UrKEkK68ZHWylCdQoEboRGpTEJGyVtNwfvP2tPDAr0bo5nSlICJJldOT0CnP/HXF3YySQhu3qu9Udiz4KDbdrk9X9ps/JOMlsUgm7TpayjaFxiehGx98a3wSGijZxCDRqPqoDUtMCAA7FnzEqr5T014Si0Rx2awhtOvYdB9qbGSulCehM1l/xRxWtn+MlTaJle0fY/0Vc4odUt7pSqENS0wIieVKANJaqe4yqpgnodNcca+/Yg6bfv/OzrLtHpv+1B3HFijAwlNSaANWHvBnWPnJzoKajtSsOKd4AUnF67JvddCJXpJyKJ/2hnTP9Kxsn7yxedOEd5UUJH+aJQSAlZ8E5SJFMuCKw5q0KcDOJ6HLrb0h5RX39hSNzXHl5XhCp6RQAGl3nMSEEPvQJ7Tr0zVpFVK7Pl3zEKXITumehH7kSy+mbG8oxaSQUpUlTwxV4fMNaU7oalacw5pTZ7B1+vuxWR0G70O3aafkMeDcUFLIs0w7Tjr7zR+S+u4jkTw7ZMgBSQ/yldLe0Gn0QU3bFOLKgbQndIkJAWDr9PdZc+oMuk07JWivmPBukHSqjE6jD2ozVVJKCvmWZseJQglA2ppM7Q2tMevmebz15DJ8B1g7+My5PTjxB/1avd6WaDxIt+TgnZgQ4sszNWAX+wqjYLekmtkZZrbQzBaZ2Q+SzO9oZn8K579mZr3yEUfDxCWs7vU0K9tNYnWvp2mYuCSr+TlX0zG7cpEiy1fPq7Nunsebk4OEAOA74M3Jy5h187xWrbc1PnXHsdRsO58aH0bNtvNzcja/acK7KcvTXWE0yvcxqiBJwcyqgPHAEKAPcKGZ9UlY7DJgvbsfAvwauCXXcTRMXMKG0fVsX9IADtuXNLBhdH3sj5ppfj7UrDineQIog8YqKV+HDDmAk37Yjy77VYNBl/2qOemH/VrdnvDWk8uyKi+6lp7QpWnATneFAYU5RhWq+mggsMjd3wUws0eBc4AFccucA4wN3z8O/M7MzHPYjetH183DG7Y3KfOG7Xx03Tw6X3xgxvktUtMxeVVR3I6jBCClJlV7Q2v4juzKi61mxTkpbyJJdsYPQVXQ1hlrUjdgp0oYobwcoxIUqvroACA+3S8Py5Iu4+7bgA3AXokrMrPRZlZvZvVr1qzJKojtSxvSlmea3xK6EhCJxlIcjVKVtwU1K84JqpYaX+Hvutu0U+gweJ8myza2DcQaqhOkKo+Xj2NUopJraHb3CcAECMZTyOazVT07B5ddScqjzG8pJQCRzD5zbg/enNy8qugz5/YoQjStl6pxOF0D9ra3P0p5hQH5O0bFK1QOXgHE/892D8uSLmNm7YHdgXW5DKLrTf2wzlVNyqxzFV1v6hdpvojkz4k/6Mfh5/WIXRlYOzj8vOLdfZRPqRqw011hQGGOUQUZeS08yL8NDCY4+NcBF7n7/LhlrgT6ufu3zGw48BX39MMitWTktYaJS/jounlsX9pAVc/OdL2pX5O6uEzzRUSKKRfHqHQjrxVsOE4zOxO4DagC7nX3m8zsRqDe3aeYWTXwIHA08AEwvLFhOpVKHo5TRKSl2sRwnO7+LPBsQtn1ce83A+r2U0SkiNpwu76IiBSakoKIiMQoKYiISIySgoiIxBTs7qN8MLM1QLadfuwNrM1DOKVA216ZtO2VJ9N2H+ju3ZLNKOmk0BJmVp/qVqxyp23XtleaSt321my3qo9ERCRGSUFERGIqMSlMKHYARaRtr0za9srT4u2uuDYFERFJrRKvFEREJIWKSwpmdquZvWVmb5jZk2a2R7FjKhQzO9/M5pvZDjOriDsyMo0NXq7M7F4ze9/M/lnsWArNzHqY2UtmtiDc368qdkyFYmbVZvZ3M3s93PafZLuOiksKwAvAEe7en6A772uLHE8h/RP4CjCz2IEUQsSxwcvV/cAZxQ6iSLYBV7t7H+CzwJUV9P/+CTDI3Y8EjgLOMLPPZrOCiksK7v58ONwnwGyCAX8qgru/6e4Lix1HAcXGBnf3LUDj2OBlz91nEnRBX3Hc/T13/0f4/iPgTZoP/1uWPLAxnOwQvrJqOK64pJDgG8DUYgcheRNlbHApY2bWi2CMlteKHErBmFmVmc0F3gdecPestr3kxmiOwsymAfslmXWdu/85XOY6gsvMiYWMLd+ibLtIJTCzLsBk4Lvu/mGx4ykUd98OHBW2lz5pZke4e+S2pbJMCu5+arr5ZjYSOBsY7GV2T26mba8wUcYGlzJkZh0IEsJEd3+i2PEUg7v/28xeImhbipwUKq76yMzOAL4PDHX3hmLHI3lVBxxqZr3NbBdgODClyDFJnpmZAfcAb7r7r4odTyGZWbfGOyrNrBNwGvBWNuuouKQA/A7oCrxgZnPN7M5iB1QoZnaumS0HjgeeMbPnih1TPoU3FIwBniNobJzk7vOLG1VhmNkjwN+Aw8xsuZldVuyYCuhzwCXAoPA3PjccI74S7A+8ZGZvEJwUveDuT2ezAj3RLCIiMZV4pSAiIikoKYiISIySgoiIxCgpiIhIjJKCiIjEKCmIiEiMkoJIC5jZDDNbb2Ydix2LSC4pKYhkKexk7SSC3ieHFjcakdxSUhDJ3qUE3a7fD4xoLDSzvczsf83sQzOrM7OfmdmsuPmfMbMXzOyDcOCfYYUPXSS9suwQTyTPLgV+RdAd82wz29fdVxMM6PMxQS+1vQi611gCYGa7EgzwdD3BoD/9CLpa+ae7Lyj4FoikoCsFkSyY2YnAgQT9KM0B3gEuCkd5Ow+4wd0bwgP9H+M+ejaw2N3vc/dt7v5/BL14nl/gTRBJS0lBJDsjgOfdfW04/XBY1o3gyjt+UJ/49wcCx5nZvxtfwMUkH/tCpGhUfSQSUdgV8TCgysxWhcUdgT2AfQkGbepOMPY3NB3LYRnwsrufVphoRVpGvaSKRGRmFxK0GxwFbImbNYmgm+LuwHbgcqAn8Dyw1N1PNLOuBAOd/IhgrGjC9Wx09zcLEb9IFKo+EoluBHCfuy9191WNL4IxOi4mGLthd2AV8CDwCPAJxAaQP51goJ+V4TK3EFxpiLQZulIQyRMzuwXYz91HZFxYpI3QlYJIjoTPIfS3wEDgMuDJYsclkg01NIvkTleCKqMaYDXwP8CfixqRSJZUfSQiIjGqPhIRkRglBRERiVFSEBGRGCUFERGJUVIQEZEYJQUREYn5/1EhokCuVl50AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Clicked std 1.019267520705244\n",
      "not clicked std 0.7063146454191541\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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Clicked std 1.0461990555400822\n",
      "not clicked std 0.6704761762169467\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"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Clicked std 0.7239588576012945\n",
      "not clicked std 0.592057888773502\n"
     ]
    }
   ],
   "source": [
    "for column in advertising:\n",
    "    if column == TARGET:\n",
    "        continue \n",
    "        \n",
    "    clicked_characteristic = advertising.loc[(advertising[TARGET] == TRUE), column]\n",
    "    clicked_mean = clicked_characteristic.mean()\n",
    "    clicked_std = clicked_characteristic.std()\n",
    "\n",
    "    not_clicked_characteristic = advertising.loc[(advertising[TARGET] == FALSE), column]\n",
    "    not_clicked_mean = not_clicked_characteristic.mean()\n",
    "    not_clicked_std = not_clicked_characteristic.std()\n",
    "\n",
    "    clicked_pdf = norm.pdf(clicked_characteristic, clicked_mean, not_clicked_std)\n",
    "    not_clicked_pdf = norm.pdf(not_clicked_characteristic, not_clicked_mean, not_clicked_std)\n",
    "    \n",
    "    fig = plt.figure()\n",
    "    plt.scatter(clicked_characteristic, clicked_pdf, c = '#e500a9')\n",
    "    plt.scatter(not_clicked_characteristic, not_clicked_pdf, c = '#9a34b5')\n",
    "    fig.suptitle('Probabilty Density Function of ' + column , fontsize = 15)\n",
    "    plt.xlabel(column, fontsize = 12)\n",
    "    plt.ylabel('Probabilty Density', fontsize = 12)\n",
    "    plt.legend(['Clicked', 'Not Clicked'])\n",
    "    plt.show()\n",
    "    print('Clicked std', clicked_std)\n",
    "    print('not clicked std', not_clicked_std)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Selecting the Best Characterstic \n",
    "<ul>\n",
    "    <li>\n",
    "The area between each two curves indicates records for which the target value will be predicted incorrectly (false negative or false positive). The selected characteristic should minimizes this area. \n",
    "    </li>\n",
    "    <li>\n",
    "The smaller the standard deviation is, the more accurate our classification is. \n",
    "    </li>\n",
    "    <li>\n",
    "The further the two curves are, the better our classification is.\n",
    "    </li>\n",
    "</ul>\n",
    "<i>Daily Internet Usage</i> is the best characteristic that satifies the criterias that are mentioned above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "selected = 'Daily Internet Usage'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Storing Mean & Std of Selected Characterstic "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "clicked_selected = advertising.loc[(advertising[TARGET] == TRUE), selected]\n",
    "clicked_mean = clicked_selected.mean()\n",
    "clicked_std = clicked_selected.std()\n",
    "\n",
    "not_clicked_selected = advertising.loc[(advertising[TARGET] == FALSE), selected]\n",
    "not_clicked_mean = not_clicked_selected.mean()\n",
    "not_clicked_std = not_clicked_selected.std()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Filling Missing Values in Target Column\n",
    "In this part missing values of target column are fill out based on the model that we designed in last parts."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "condition = (advertising[selected] < clicked_mean + clicked_std) & \\\n",
    "    (advertising[selected] > clicked_mean - clicked_std)\n",
    "\n",
    "advertising = advertising[pd.isnull(advertising[TARGET])]\n",
    "\n",
    "advertising[TARGET] = advertising[TARGET].fillna(condition.map({True: CLICKED, False: NOT_CLICKED}))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Storing Prediction Result in a File\n",
    "In this part the prediction result was written in the *prediction_result.csv* file using Pandas *to_csv* method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "advertising[TARGET].to_csv('prediction_result.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Daily Time Spent on Site</th>\n",
       "      <th>Age</th>\n",
       "      <th>Area Income</th>\n",
       "      <th>Daily Internet Usage</th>\n",
       "      <th>Clicked on Ad</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>-1.129048</td>\n",
       "      <td>1.494773</td>\n",
       "      <td>-0.698304</td>\n",
       "      <td>-1.374546</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>-1.437721</td>\n",
       "      <td>-0.346249</td>\n",
       "      <td>-1.790880</td>\n",
       "      <td>-0.861692</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>-0.088509</td>\n",
       "      <td>-1.496888</td>\n",
       "      <td>-0.210052</td>\n",
       "      <td>-0.931215</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>0.817766</td>\n",
       "      <td>-0.691441</td>\n",
       "      <td>0.730719</td>\n",
       "      <td>1.064011</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>-1.540393</td>\n",
       "      <td>0.574262</td>\n",
       "      <td>1.040827</td>\n",
       "      <td>-0.298362</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Daily Time Spent on Site       Age  Area Income  Daily Internet Usage  \\\n",
       "10                 -1.129048  1.494773    -0.698304             -1.374546   \n",
       "14                 -1.437721 -0.346249    -1.790880             -0.861692   \n",
       "15                 -0.088509 -1.496888    -0.210052             -0.931215   \n",
       "20                  0.817766 -0.691441     0.730719              1.064011   \n",
       "24                 -1.540393  0.574262     1.040827             -0.298362   \n",
       "\n",
       "    Clicked on Ad  \n",
       "10            1.0  \n",
       "14            1.0  \n",
       "15            1.0  \n",
       "20            0.0  \n",
       "24            1.0  "
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "advertising.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conclusion\n",
    "In this computer assignment we learned that Python\n",
    "libraries are very powerful tools in Artificial Intelligence. Also\n",
    "we were introduced one of the applications of Artificial\n",
    "Intelligence in real life."
   ]
  }
 ],
 "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
}