Customer_habit_analysis_model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Project description\n",
    "Mobile Operator Megaline was dissatisfied because most of their customers still use\n",
    "the old package. The company wants to develop a model that can analyze consumer behavior and recommend one of Megaline's two new plans: Smart or Ultra.\n",
    "\n",
    "You have access to behavioral data of customers who have switched to the latest package (from the Statistical Data Analysis course project). In this classification task, you need to develop a model that is able to choose the right package. Well, since you have completed the data pre-processing step, you can proceed directly to the model creation stage.\n",
    "\n",
    "Develop a model with the highest level of accuracy. In this project, the threshold for accuracy is 0.75. Don't forget to check the accuracy of your model using the test dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load required libraries\n",
    "\n",
    "#data prep\n",
    "import pandas as pd, numpy as np\n",
    "\n",
    "#data viz\n",
    "import matplotlib.pyplot as plt, seaborn as sns\n",
    "\n",
    "#skylearn\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.metrics import accuracy_score\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Pra-process data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load data file to DataFrame\n",
    "try:\n",
    "    data = pd.read_csv ('/datasets/users_behavior.csv')\n",
    "except:\n",
    "    data = pd.read_csv ('users_behavior.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<b>Deskripsi data</b>\n",
    "\n",
    "Each observation in our dataset contains monthly behavioral information about a single user. The information includes:\n",
    "\n",
    "<b>сalls</b> — Total Inbound Call\n",
    "\n",
    "<b>minutes</b> — Total Duration Call (Minute)\n",
    "\n",
    "<b>messages</b> — Total spend Messages\n",
    "\n",
    "<b>mb_used</b> — traffic internet uses (MB-Mega Byte)\n",
    "\n",
    "<b>is_ultimate</b> — running packages in current month (Ultimate - 1, Surf - 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Exploration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 3214 entries, 0 to 3213\n",
      "Data columns (total 5 columns):\n",
      " #   Column    Non-Null Count  Dtype  \n",
      "---  ------    --------------  -----  \n",
      " 0   calls     3214 non-null   float64\n",
      " 1   minutes   3214 non-null   float64\n",
      " 2   messages  3214 non-null   float64\n",
      " 3   mb_used   3214 non-null   float64\n",
      " 4   is_ultra  3214 non-null   int64  \n",
      "dtypes: float64(4), int64(1)\n",
      "memory usage: 125.7 KB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>calls</th>\n",
       "      <th>minutes</th>\n",
       "      <th>messages</th>\n",
       "      <th>mb_used</th>\n",
       "      <th>is_ultra</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>40.0</td>\n",
       "      <td>311.90</td>\n",
       "      <td>83.0</td>\n",
       "      <td>19915.42</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>85.0</td>\n",
       "      <td>516.75</td>\n",
       "      <td>56.0</td>\n",
       "      <td>22696.96</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>77.0</td>\n",
       "      <td>467.66</td>\n",
       "      <td>86.0</td>\n",
       "      <td>21060.45</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>106.0</td>\n",
       "      <td>745.53</td>\n",
       "      <td>81.0</td>\n",
       "      <td>8437.39</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>66.0</td>\n",
       "      <td>418.74</td>\n",
       "      <td>1.0</td>\n",
       "      <td>14502.75</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>58.0</td>\n",
       "      <td>344.56</td>\n",
       "      <td>21.0</td>\n",
       "      <td>15823.37</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>57.0</td>\n",
       "      <td>431.64</td>\n",
       "      <td>20.0</td>\n",
       "      <td>3738.90</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>15.0</td>\n",
       "      <td>132.40</td>\n",
       "      <td>6.0</td>\n",
       "      <td>21911.60</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>7.0</td>\n",
       "      <td>43.39</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2538.67</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>90.0</td>\n",
       "      <td>665.41</td>\n",
       "      <td>38.0</td>\n",
       "      <td>17358.61</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   calls  minutes  messages   mb_used  is_ultra\n",
       "0   40.0   311.90      83.0  19915.42         0\n",
       "1   85.0   516.75      56.0  22696.96         0\n",
       "2   77.0   467.66      86.0  21060.45         0\n",
       "3  106.0   745.53      81.0   8437.39         1\n",
       "4   66.0   418.74       1.0  14502.75         0\n",
       "5   58.0   344.56      21.0  15823.37         0\n",
       "6   57.0   431.64      20.0   3738.90         1\n",
       "7   15.0   132.40       6.0  21911.60         0\n",
       "8    7.0    43.39       3.0   2538.67         1\n",
       "9   90.0   665.41      38.0  17358.61         0"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<pre><b>Mini-Conclusion</b>\n",
    "\n",
    "<b>сalls</b> — is <b>Interger</b>, Need to do something.\n",
    "\n",
    "<b>minutes</b> — is OK\n",
    "\n",
    "<b>messages</b> — is <b>Interger</b>, Need to do something.\n",
    "\n",
    "<b>mb_used</b> — is OK\n",
    "\n",
    "<b>is_ultimate</b> — is OK\n",
    "</pre>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Quality data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Change data type according to requirement\n",
    "data.calls = data.calls.astype(int)\n",
    "data.messages = data.messages.astype(int)\n",
    "data.mb_used = data.mb_used.astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 3214 entries, 0 to 3213\n",
      "Data columns (total 5 columns):\n",
      " #   Column    Non-Null Count  Dtype  \n",
      "---  ------    --------------  -----  \n",
      " 0   calls     3214 non-null   int64  \n",
      " 1   minutes   3214 non-null   float64\n",
      " 2   messages  3214 non-null   int64  \n",
      " 3   mb_used   3214 non-null   int64  \n",
      " 4   is_ultra  3214 non-null   int64  \n",
      "dtypes: float64(1), int64(4)\n",
      "memory usage: 125.7 KB\n"
     ]
    }
   ],
   "source": [
    "#Checking data\n",
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking duplicate data\n",
    "data.duplicated().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "calls       0\n",
       "minutes     0\n",
       "messages    0\n",
       "mb_used     0\n",
       "is_ultra    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking missing value\n",
    "data.isna().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<pre><b>Mini-Conclusion</b>\n",
    "\n",
    "Quality data is OK\n",
    "</pre>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploratory Data Analyst"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# separating registered plans\n",
    "data['plan'] = data['is_ultra']\n",
    "data.loc[(data['is_ultra'] == 1), 'plan'] = 'ultimate'\n",
    "data.loc[(data['is_ultra'] == 0), 'plan'] = 'surf'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Sliced data based on the plan user used\n",
    "surf_user = data.loc[data['plan'] == 'surf']\n",
    "ultimate_user = data.loc[data['plan'] == 'ultimate']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show distribution count of call usage based on plan\n",
    "sns.violinplot(data=data, x='calls', y='plan')\n",
    "plt.title('Total counting call distribution based on plan')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>var</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>calls</th>\n",
       "      <th>calls</th>\n",
       "      <th>calls</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>plan</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>surf</th>\n",
       "      <td>58.463437</td>\n",
       "      <td>672.876243</td>\n",
       "      <td>25.939858</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ultimate</th>\n",
       "      <td>73.392893</td>\n",
       "      <td>1928.689990</td>\n",
       "      <td>43.916853</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               mean          var        std\n",
       "              calls        calls      calls\n",
       "plan                                       \n",
       "surf      58.463437   672.876243  25.939858\n",
       "ultimate  73.392893  1928.689990  43.916853"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# getting Mean, variant dan standar deviation values based on plan\n",
    "data.pivot_table(index='plan', values='calls', aggfunc=[np.mean, np.var, np.std])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Insight.</b>\n",
    "\n",
    "- <b>Ultimate</b> plan have more spread of data and a higher mean value compared with <b>Surf</b> plan\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show the distribution of call duration based on plan\n",
    "sns.violinplot(data=data, x='minutes', y='plan')\n",
    "plt.title('Call duration distribution based on plan')\n",
    "plt.show()"
   ]
  },
  {
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>var</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>minutes</th>\n",
       "      <th>minutes</th>\n",
       "      <th>minutes</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>plan</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>surf</th>\n",
       "      <td>405.942952</td>\n",
       "      <td>34044.901170</td>\n",
       "      <td>184.512604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ultimate</th>\n",
       "      <td>511.224569</td>\n",
       "      <td>94883.158571</td>\n",
       "      <td>308.031100</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                mean           var         std\n",
       "             minutes       minutes     minutes\n",
       "plan                                          \n",
       "surf      405.942952  34044.901170  184.512604\n",
       "ultimate  511.224569  94883.158571  308.031100"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# getting Mean, variant dan standar deviation values based on plan\n",
    "data.pivot_table(index='plan', values='minutes', aggfunc=[np.mean, np.var, np.std])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Insight.</b>\n",
    "\n",
    "- <b>Ultimate</b> plan have more spread of data and a higher mean value compared with <b>Surf</b> plan\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show distribution of messages based on plan\n",
    "sns.violinplot(data=data, x='messages', y='plan')\n",
    "plt.title('Messages distribution based on plan')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>var</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>messages</th>\n",
       "      <th>messages</th>\n",
       "      <th>messages</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>plan</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>surf</th>\n",
       "      <td>33.384029</td>\n",
       "      <td>796.812958</td>\n",
       "      <td>28.227876</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ultimate</th>\n",
       "      <td>49.363452</td>\n",
       "      <td>2285.266143</td>\n",
       "      <td>47.804457</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               mean          var        std\n",
       "           messages     messages   messages\n",
       "plan                                       \n",
       "surf      33.384029   796.812958  28.227876\n",
       "ultimate  49.363452  2285.266143  47.804457"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# getting Mean, variant dan standar deviation values based on plan\n",
    "data.pivot_table(index='plan', values='messages', aggfunc=[np.mean, np.var, np.std])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Insight.</b>\n",
    "\n",
    "- <b>Ultimate</b> plan have more spread of data and a higher mean value compared with <b>Surf</b> plan\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Show distribution of internet uses based on plan\n",
    "sns.violinplot(data=data, x='mb_used', y='plan')\n",
    "plt.title('internet distribution based on plan')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>var</th>\n",
       "      <th>std</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>mb_used</th>\n",
       "      <th>mb_used</th>\n",
       "      <th>mb_used</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>plan</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>surf</th>\n",
       "      <td>16207.976223</td>\n",
       "      <td>3.446283e+07</td>\n",
       "      <td>5870.504880</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ultimate</th>\n",
       "      <td>19468.344162</td>\n",
       "      <td>1.017512e+08</td>\n",
       "      <td>10087.177813</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                  mean           var           std\n",
       "               mb_used       mb_used       mb_used\n",
       "plan                                              \n",
       "surf      16207.976223  3.446283e+07   5870.504880\n",
       "ultimate  19468.344162  1.017512e+08  10087.177813"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# getting Mean, variant dan standar deviation values based on plan\n",
    "data.pivot_table(index='plan', values='mb_used', aggfunc=[np.mean, np.var, np.std])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Insight.</b>\n",
    "\n",
    "- <b>Ultimate</b> plan have more spread of data and a higher mean value compared with <b>Surf</b> plan\n",
    "    \n",
    "    \n",
    "    \n",
    "<b>Mini-Conclusion.</b>\n",
    "    \n",
    "- <b>Ultimate</b> plan on any data distribution have more spread of data and a higher mean value compared with <b>Surf</b> plan\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Feature Engineering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Separating target and feature (X is feature, y is target)\n",
    "X =  data.drop(['is_ultra','plan'], axis=1)\n",
    "y = data['is_ultra']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split data into Train and test\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split test data into validation and test\n",
    "X_val, X_test, y_val, y_test = train_test_split(X_test, y_test, test_size=0.5, random_state=25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1928, 4)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking dimension of data train \n",
    "X_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(643, 4)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking dimension of data validation \n",
    "X_val.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(643, 4)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Checking dimension of data test \n",
    "X_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Training"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Logistic Regression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression()"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Define Model\n",
    "lr = LogisticRegression()\n",
    "# Training the model\n",
    "lr.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7427385892116183"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Training prediction\n",
    "y_train_pred_lr = lr.predict(X_train)\n",
    "\n",
    "# accuracy of training\n",
    "accuracy_score(y_train, y_train_pred_lr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7511664074650077"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Validate predication\n",
    "y_val_pred_lr = lr.predict(X_val)\n",
    "\n",
    "# accuracy of validation\n",
    "accuracy_score(y_val, y_val_pred_lr)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Insight.</b>\n",
    "\n",
    "using the Logistic regression method had an accuracy in validation predict of 75%, \n",
    "that has a 1% difference with training predict.\n",
    "    \n",
    "its show that model was fit in 75% accuracy.\n",
    "    \n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Decision Tree Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier()"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Define Model\n",
    "dt = DecisionTreeClassifier()\n",
    "# Training the model\n",
    "dt.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Training prediction\n",
    "y_train_pred_dt = dt.predict(X_train)\n",
    "\n",
    "# accuracy of training\n",
    "accuracy_score(y_train, y_train_pred_dt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7340590979782271"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Validate predication\n",
    "y_val_pred_dt = dt.predict(X_val)\n",
    "\n",
    "# accuracy of validation\n",
    "accuracy_score(y_val, y_val_pred_dt)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Insight.</b>\n",
    "\n",
    "using the Decision Tree method had an accuracy in validation predict of 73%, \n",
    "that has a 27% difference with training predict.\n",
    "    \n",
    "it show that model was overfit. then we must tuning the parameter in decision tree method.\n",
    "    \n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Hyperparameter Tuning Decision Tree"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(max_depth=6, min_samples_leaf=5, min_samples_split=5,\n",
       "                       random_state=100)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Define Model\n",
    "dt = DecisionTreeClassifier(random_state=100,max_depth=6, min_samples_split=5, min_samples_leaf=5)\n",
    "# Training the model\n",
    "dt.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8298755186721992"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Training prediction\n",
    "y_train_pred_dt = dt.predict(X_train)\n",
    "\n",
    "# accuracy of training\n",
    "accuracy_score(y_train, y_train_pred_dt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7931570762052877"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Validate predication\n",
    "y_val_pred_dt = dt.predict(X_val)\n",
    "\n",
    "# accuracy of validation\n",
    "accuracy_score(y_val, y_val_pred_dt)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Insight.</b>\n",
    "\n",
    "After using the hyperparameter tuning Decision Tree method had an accuracy in validation predict of 79.3%, \n",
    "that has a 3.6% difference with training predict.\n",
    "    \n",
    "it show that model was fit.\n",
    "    \n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forest Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier()"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Define Model\n",
    "rf = RandomForestClassifier()\n",
    "# Training the model\n",
    "rf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Training prediction\n",
    "y_train_pred_rf = rf.predict(X_train)\n",
    "\n",
    "# accuracy of training\n",
    "accuracy_score(y_train, y_train_pred_rf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7978227060653188"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Validate predication\n",
    "y_val_pred_rf = rf.predict(X_val)\n",
    "\n",
    "# accuracy of validation\n",
    "accuracy_score(y_val, y_val_pred_rf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Insight.</b>\n",
    "\n",
    "using the Random Forrest method had an accuracy in validation predict of 79%, \n",
    "that has a 21% difference with training predict.\n",
    "    \n",
    "it show that model was overfit. then we must tuning the parameter in random forest method.\n",
    "    \n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Hyperparameter Tuning Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(max_depth=5, n_estimators=10)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Define Model\n",
    "rf = RandomForestClassifier(n_estimators=10, max_depth=5)\n",
    "# Training the model\n",
    "rf.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8241701244813278"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Training prediction\n",
    "y_train_pred_rf = rf.predict(X_train)\n",
    "\n",
    "# accuracy of training\n",
    "accuracy_score(y_train, y_train_pred_rf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7978227060653188"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Validate predication\n",
    "y_val_pred_rf = rf.predict(X_val)\n",
    "\n",
    "# accuracy of validation\n",
    "accuracy_score(y_val, y_val_pred_rf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Insight.</b>\n",
    "\n",
    "After using the hyperparamater tuning Random Forrest method had an accuracy in validation predict of 79.3%, \n",
    "that has a 2.9% difference with training predict.\n",
    "    \n",
    "it show that model was fit.\n",
    "    \n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Mini-Conclusion.</b>\n",
    "\n",
    "The best method who has big accuracy with a slight difference is using Hyperparameter tuning random Forest\n",
    "    \n",
    "\n",
    "</div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predict data testing using the best model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Hyperparameter Tuning Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8258164852255054"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Prediksi di Testing\n",
    "y_test_pred_rf = rf.predict(X_test)\n",
    "\n",
    "# Cek Akurasi di Testing\n",
    "accuracy_score(y_test, y_test_pred_rf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert-info\">\n",
    "<b>Insight.</b>\n",
    "\n",
    "After using the hyperparamater tuning Random Forrest method had an accuracy in validation predict of 79.3%, \n",
    "that has a 2.9% difference with training predict.\n",
    "    \n",
    "it show that model was fit.\n",
    "    \n",
    "\n",
    "</div>"
   ]
  }
 ],
 "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": "3.9.19"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": true,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "321.333px"
   },
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}