Prudential_Code.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Importing the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def readdata(path):\n",
    "    rawdata = pd.read_csv(path)\n",
    "    return rawdata\n",
    "\n",
    "rawdata = readdata('Rawdata/train.csv')\n",
    "\n",
    "#rawdata = pd.read_csv()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 59381 entries, 0 to 59380\n",
      "Columns: 128 entries, Id to Response\n",
      "dtypes: float64(18), int64(109), object(1)\n",
      "memory usage: 58.0+ MB\n"
     ]
    }
   ],
   "source": [
    "rawdata.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>Id</th>\n",
       "      <th>Product_Info_1</th>\n",
       "      <th>Product_Info_2</th>\n",
       "      <th>Product_Info_3</th>\n",
       "      <th>Product_Info_4</th>\n",
       "      <th>Product_Info_5</th>\n",
       "      <th>Product_Info_6</th>\n",
       "      <th>Product_Info_7</th>\n",
       "      <th>Ins_Age</th>\n",
       "      <th>Ht</th>\n",
       "      <th>...</th>\n",
       "      <th>Medical_Keyword_40</th>\n",
       "      <th>Medical_Keyword_41</th>\n",
       "      <th>Medical_Keyword_42</th>\n",
       "      <th>Medical_Keyword_43</th>\n",
       "      <th>Medical_Keyword_44</th>\n",
       "      <th>Medical_Keyword_45</th>\n",
       "      <th>Medical_Keyword_46</th>\n",
       "      <th>Medical_Keyword_47</th>\n",
       "      <th>Medical_Keyword_48</th>\n",
       "      <th>Response</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>D3</td>\n",
       "      <td>10</td>\n",
       "      <td>0.076923</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.641791</td>\n",
       "      <td>0.581818</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>A1</td>\n",
       "      <td>26</td>\n",
       "      <td>0.076923</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0.059701</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>E1</td>\n",
       "      <td>26</td>\n",
       "      <td>0.076923</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0.029851</td>\n",
       "      <td>0.745455</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>D4</td>\n",
       "      <td>10</td>\n",
       "      <td>0.487179</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0.164179</td>\n",
       "      <td>0.672727</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>D2</td>\n",
       "      <td>26</td>\n",
       "      <td>0.230769</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0.417910</td>\n",
       "      <td>0.654545</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 128 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Id  Product_Info_1 Product_Info_2  Product_Info_3  Product_Info_4  \\\n",
       "0   2               1             D3              10        0.076923   \n",
       "1   5               1             A1              26        0.076923   \n",
       "2   6               1             E1              26        0.076923   \n",
       "3   7               1             D4              10        0.487179   \n",
       "4   8               1             D2              26        0.230769   \n",
       "\n",
       "   Product_Info_5  Product_Info_6  Product_Info_7   Ins_Age        Ht  \\\n",
       "0               2               1               1  0.641791  0.581818   \n",
       "1               2               3               1  0.059701  0.600000   \n",
       "2               2               3               1  0.029851  0.745455   \n",
       "3               2               3               1  0.164179  0.672727   \n",
       "4               2               3               1  0.417910  0.654545   \n",
       "\n",
       "     ...     Medical_Keyword_40  Medical_Keyword_41  Medical_Keyword_42  \\\n",
       "0    ...                      0                   0                   0   \n",
       "1    ...                      0                   0                   0   \n",
       "2    ...                      0                   0                   0   \n",
       "3    ...                      0                   0                   0   \n",
       "4    ...                      0                   0                   0   \n",
       "\n",
       "   Medical_Keyword_43  Medical_Keyword_44  Medical_Keyword_45  \\\n",
       "0                   0                   0                   0   \n",
       "1                   0                   0                   0   \n",
       "2                   0                   0                   0   \n",
       "3                   0                   0                   0   \n",
       "4                   0                   0                   0   \n",
       "\n",
       "   Medical_Keyword_46  Medical_Keyword_47  Medical_Keyword_48  Response  \n",
       "0                   0                   0                   0         8  \n",
       "1                   0                   0                   0         4  \n",
       "2                   0                   0                   0         8  \n",
       "3                   0                   0                   0         8  \n",
       "4                   0                   0                   0         8  \n",
       "\n",
       "[5 rows x 128 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rawdata.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "_____________________________________________________________________________________________________________________________________________________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Finding missing values in the training dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20b55086898>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def missing_value_plot(data):\n",
    "    plt.figure(figsize=(20, 10))\n",
    "    nv = sns.heatmap(data.isnull(), yticklabels=False,cbar=False,cmap='viridis')\n",
    "    nv.set_xticklabels(\n",
    "        nv.get_xticklabels(),\n",
    "        rotation=45\n",
    "    );\n",
    "    \n",
    "    return nv\n",
    "\n",
    "missing_value_plot(rawdata)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Seaborn helps to visualize if there are null values present in the dataset. Now that we have figured out there are, we need to find those features and check if they are salvagable through data augmentation or have to get rid of it."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Listing out columns with null values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Employment_Info_1', 'Employment_Info_4', 'Employment_Info_6',\n",
       "       'Insurance_History_5', 'Family_Hist_2', 'Family_Hist_3',\n",
       "       'Family_Hist_4', 'Family_Hist_5', 'Medical_History_1',\n",
       "       'Medical_History_10', 'Medical_History_15', 'Medical_History_24',\n",
       "       'Medical_History_32'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rawdata.columns[rawdata.isnull().any()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4\n"
     ]
    }
   ],
   "source": [
    "a= [1,2,3,4,5,6,7]\n",
    "from statistics import mean\n",
    "print(mean(a))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we can see that columns listed may have various degree of missing values and might be relevant to predict the outcome. So we need to find columns who have more than 50% missing values so that they can be eliminated as they won't help the model predict the outcome in the way other columns could."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Below are the list of columns that have more than 50% of missing values and will be dropped from the main training dataset as augmenting the data will be less useful for the correctness of predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Employment_Info_4', 'Employment_Info_6', 'Insurance_History_5',\n",
       "       'Family_Hist_2', 'Family_Hist_3', 'Family_Hist_4', 'Family_Hist_5',\n",
       "       'Medical_History_1', 'Medical_History_10', 'Medical_History_15',\n",
       "       'Medical_History_24', 'Medical_History_32'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rawdata.columns[rawdata.isna().mean().round(4) * 100 > .5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "rawdata = rawdata.drop(['Employment_Info_4', 'Employment_Info_6', 'Insurance_History_5',\n",
    "       'Family_Hist_2', 'Family_Hist_3', 'Family_Hist_4', 'Family_Hist_5',\n",
    "       'Medical_History_1', 'Medical_History_10', 'Medical_History_15',\n",
    "       'Medical_History_24', 'Medical_History_32'], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 59381 entries, 0 to 59380\n",
      "Columns: 116 entries, Id to Response\n",
      "dtypes: float64(6), int64(109), object(1)\n",
      "memory usage: 52.6+ MB\n"
     ]
    }
   ],
   "source": [
    "rawdata.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we have 116 features in the dataset after dropping the columns with more than 50% missing values. Let's check how many columns we have with missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Employment_Info_1'], dtype='object')"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rawdata.columns[rawdata.isnull().any()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is only one column has null values now which is 'Employment_Info_1'. As they are integers, they can be replaced by the mean values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "rawdata['Employment_Info_1'].fillna((rawdata['Employment_Info_1'].mean()), inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index([], dtype='object')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rawdata.columns[rawdata.isnull().any()]"
   ]
  },
  {
   "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>Id</th>\n",
       "      <th>Product_Info_1</th>\n",
       "      <th>Product_Info_2</th>\n",
       "      <th>Product_Info_3</th>\n",
       "      <th>Product_Info_4</th>\n",
       "      <th>Product_Info_5</th>\n",
       "      <th>Product_Info_6</th>\n",
       "      <th>Product_Info_7</th>\n",
       "      <th>Ins_Age</th>\n",
       "      <th>Ht</th>\n",
       "      <th>...</th>\n",
       "      <th>Medical_Keyword_40</th>\n",
       "      <th>Medical_Keyword_41</th>\n",
       "      <th>Medical_Keyword_42</th>\n",
       "      <th>Medical_Keyword_43</th>\n",
       "      <th>Medical_Keyword_44</th>\n",
       "      <th>Medical_Keyword_45</th>\n",
       "      <th>Medical_Keyword_46</th>\n",
       "      <th>Medical_Keyword_47</th>\n",
       "      <th>Medical_Keyword_48</th>\n",
       "      <th>Response</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>D3</td>\n",
       "      <td>10</td>\n",
       "      <td>0.076923</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.641791</td>\n",
       "      <td>0.581818</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>A1</td>\n",
       "      <td>26</td>\n",
       "      <td>0.076923</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0.059701</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>E1</td>\n",
       "      <td>26</td>\n",
       "      <td>0.076923</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0.029851</td>\n",
       "      <td>0.745455</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>D4</td>\n",
       "      <td>10</td>\n",
       "      <td>0.487179</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0.164179</td>\n",
       "      <td>0.672727</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "      <td>D2</td>\n",
       "      <td>26</td>\n",
       "      <td>0.230769</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0.417910</td>\n",
       "      <td>0.654545</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 116 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Id  Product_Info_1 Product_Info_2  Product_Info_3  Product_Info_4  \\\n",
       "0   2               1             D3              10        0.076923   \n",
       "1   5               1             A1              26        0.076923   \n",
       "2   6               1             E1              26        0.076923   \n",
       "3   7               1             D4              10        0.487179   \n",
       "4   8               1             D2              26        0.230769   \n",
       "\n",
       "   Product_Info_5  Product_Info_6  Product_Info_7   Ins_Age        Ht  \\\n",
       "0               2               1               1  0.641791  0.581818   \n",
       "1               2               3               1  0.059701  0.600000   \n",
       "2               2               3               1  0.029851  0.745455   \n",
       "3               2               3               1  0.164179  0.672727   \n",
       "4               2               3               1  0.417910  0.654545   \n",
       "\n",
       "     ...     Medical_Keyword_40  Medical_Keyword_41  Medical_Keyword_42  \\\n",
       "0    ...                      0                   0                   0   \n",
       "1    ...                      0                   0                   0   \n",
       "2    ...                      0                   0                   0   \n",
       "3    ...                      0                   0                   0   \n",
       "4    ...                      0                   0                   0   \n",
       "\n",
       "   Medical_Keyword_43  Medical_Keyword_44  Medical_Keyword_45  \\\n",
       "0                   0                   0                   0   \n",
       "1                   0                   0                   0   \n",
       "2                   0                   0                   0   \n",
       "3                   0                   0                   0   \n",
       "4                   0                   0                   0   \n",
       "\n",
       "   Medical_Keyword_46  Medical_Keyword_47  Medical_Keyword_48  Response  \n",
       "0                   0                   0                   0         8  \n",
       "1                   0                   0                   0         4  \n",
       "2                   0                   0                   0         8  \n",
       "3                   0                   0                   0         8  \n",
       "4                   0                   0                   0         8  \n",
       "\n",
       "[5 rows x 116 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rawdata.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As there are no null values in the dataset. It's time to do some exploratory analysis to find what the data really consists of."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "___________________________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploratory Data Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We know that this is a insurance applicant data consisting of the person's medical and some of the insurance data. We can explore applicants age, height, wt and the risk factor to see how they are correlated with each other."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\mites\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6462: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n",
      "  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20b5521a4e0>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 5))\n",
    "sns.distplot(rawdata['Ht']*10, bins = 30)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This histogram shows that the 'Height' of the applicants ranges between 4ft and 10ft assuming the values indicated are the actual values rather than scaled values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\mites\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6462: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n",
      "  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20b038950f0>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 5))\n",
    "sns.distplot(rawdata['Ins_Age']*100, bins = 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The age histogram showcases the number of applicants and their age who have applied for the insurance.\n",
    "We can see that the max number of people are around the age of 25,40 and 60 years old."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's look into it in more detial by looking at the Probability and Cumilative Distribution function to undertand how many values lie in a what range."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2.03432074e-02 3.39670939e-02 6.29494283e-02 6.78163049e-02\n",
      " 7.35420421e-02 9.22348899e-02 6.11306647e-02 7.01571210e-02\n",
      " 9.24201344e-02 7.03255250e-02 7.61186238e-02 8.91699365e-02\n",
      " 6.42292989e-02 5.60617032e-02 4.29261885e-02 1.61667874e-02\n",
      " 8.36968054e-03 1.80192317e-03 2.35765649e-04 3.36808070e-05]\n",
      "[  0.   5.  10.  15.  20.  25.  30.  35.  40.  45.  50.  55.  60.  65.\n",
      "  70.  75.  80.  85.  90.  95. 100.]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x20b55ea8080>]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAD8CAYAAACMwORRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3Xl8VPW9//HXJzvZIJCFsIMGES2KRtytK+JKqyJQqbvctlp77+1yvbU/29rbe9va26qt1VKvO4qKGyqKuG8VCYIgKASQJRCSACEL2TPf3x9ngBgDGWCSM5l5Px+PYeacOZl8ZibznsP3fM/3a845REQkusT5XYCIiISfwl1EJAop3EVEopDCXUQkCincRUSikMJdRCQKKdxFRKKQwl1EJAop3EVEolCCX784OzvbDRs2zK9fLyLSIy1atGircy6ns+18C/dhw4ZRVFTk168XEemRzGx9KNupWUZEJAop3EVEopDCXUQkCincRUSikMJdRCQKdRruZvaAmZWb2Wd7ud/M7G4zW21mS83smPCXKSIi+yOUPfeHgAn7uP88oCB4mQ7ce/BliYjIwei0n7tz7l0zG7aPTSYCjzhvvr6PzKyPmeU750rDVKOIiCfQCnXboaUeWpuDlybvOrDrdhO0tuy5HWj5+noXCD6gA+e8a9hze/fso3u5vzOdTV962AQYeOx+Pvn9E46TmAYCG9sslwTXfS3czWw63t49Q4YMCcOvFpGo4Bw0VEHNFqjZ7F1XB69rSoOXLd7FtfpdbYhs73dl9O8R4d7RM+jwa8s5NwOYAVBYWKiZuUViSUM1bFkKWz6D6hKoLv1qmDfXff1nUnpDxgAvDHNGedfp/SGxF8QnQXxi8BK8HZfY8fr4pOB9wYvFsTu6zLzbtivKbM+6ju63fYR2BAlHuJcAg9ssDwI2h+FxRaSnaqiG0k+hdAlsXuJdb1u95/6EFC+oMwZA/tFwWDDAM/KDl+DtpFT/nkMPF45wnwPcZGazgOOBKrW3i8SQzoI8c6AX4GMme9f5R0F6bo/ZA+6pOg13M3sCOB3INrMS4JdAIoBz7j5gLnA+sBqoA67pqmJFxGfOwYaPYFMRbF7shfn2NXvuzxwEA46GMVO86/yjIb3TAQylC4TSW2ZqJ/c74MawVSQikWntO/DGr2HTIm95V5AfNVVBHoF8G/JXRHqITZ/AG7fD2re8JpaL7oZRF0Batt+VyT4o3EWkY1uL4c3fwIoXoFdfGP9bOO56SEzxuzIJgcJdRL6qahO88ztYPNPrcvjN/4ATb4KUTL8rk/2gcBcRz85t8P6f4ON/AA7GTYdTf6x29B5K4S4S6xpr4aO/wQd3Q/NO7wDp6bdAH51F3pMp3EViVUsjFD0I794BdVth1IVw5i8g93C/K5MwULiLxJpAKyx9Ct76b6jaAMNOhbN+CYOP87syCSOFu0gs2bgQ5vwQKj73+qVfdCcccqbOFo1CCneRWNDSBO/83jtgmjkQJj0Moycq1KOYwl0k2pV/Ac9N98Z/Oeo7cN7vvNEWJaop3EWiVSAAC+6D138Fyekw+TE4/CK/q5JuonAXiUY7NsLz34d178HI8+Diu72RGCVmKNxFoolz8OkseOVn3lRyF90Nx1yptvUYpHAXiRY7t8FLP4LPX4TBJ8C374O+w/2uSnyicBeJBqvmwQs3QX0lnP0rOOlmiIv3uyrxkcJdpCdrrIXXboVFD0HuEfDdZ6H/N/yuSiKAwl2kp9qwAJ77F6hc5+2pn/kLSEj2uyqJEAp3kZ6mpQne/h/44E5vNqSrX4ZhJ/tdlUQYhbtIT7JtDTx9NWxZCkdPgwn/o3HWpUMKd5GeYuWr8Ox0iIuDKY97U92J7IXCXSTSBQLw7h+8ppj+Y7wzTbOG+l2VRDiFu0gka6iCZ/8FVr0CY6Z4ozgm9vK7KukBFO4ikar8c5h1BexYD+fdAeNu0JmmEjKFu0gkWv48PP8DSEqDq16EoSf5XZH0MAp3kUgSaIU3bve6OQ46Di5/BDIH+F2V9EAKd5FIUbcdZl8Da9+Gwmthwu90UpIcMIW7SCQo/RRmTYPaLXDxX+GY7/pdkfRwCncRv306C178EaT2g2tfhYHH+l2RRAGFu4hfWpth3q3w8d9h2Klw2YOQnuN3VRIlFO4ifqgp84YR2PAhnHAjnHM7xOvjKOETF8pGZjbBzFaa2Wozu6WD+4eY2VtmttjMlprZ+eEvVSRKbFwIM74JmxfDJffDhP9WsEvYdRruZhYP3AOcB4wGpprZ6Hab/QJ4yjk3FpgC/C3chYpEheL58ND5EJ8E18+HMZP8rkiiVCh77uOA1c65tc65JmAWMLHdNg7YNTRdb2Bz+EoUiRIbF8JTV0LuaJj+tibVkC4Vyv8FBwIb2yyXAMe32+ZXwGtm9kMgDTg7LNWJRIuKlfD4JMjoD1fMhtS+flckUS6UPfeOBrNw7ZanAg855wYB5wOPmtnXHtvMpptZkZkVVVRU7H+1Ij1RVQk8+m2vKea7z6lHjHSLUMK9BBjcZnkQX292uQ54CsA5908gBchu/0DOuRnOuULnXGFOjv7AJQbUbYdHL4HGGpj2DGQN87siiRGhhPtCoMDMhptZEt4B0zntttkAnAVgZofjhbt2zSW2NdXB45O9OU6nPqE2dulWnYa7c64FuAmYB3yO1ytmuZndbmYXBzf7MXCDmX0KPAFc7Zxr33QjEjtam+Hpq2BTEVx6Pww7xe+KJMaE1LnWOTcXmNtu3W1tbq8ANEOvCIBzMOeHUPwaXHgnjL64858RCbOQTmISkf0w/zb49Ak441YovMbvaiRGKdxFwunDv8CHd8NxN8BpP/W7GolhCneRcPl0Frz2Cxj9LTjv95oST3ylcBcJh+L58MKNMPw0uGQGxMX7XZHEOIW7yMHaNaxA3hEweaZmT5KIoHAXORi7hhVIz/OGFUjJ7PxnRLqBwl3kQO0aViAuMTisQK7fFYnspkGkRQ5E3XZ47FJvWIGrX4a+w/2uSOQrFO4i+2vXsALb18K0ZyF/jN8ViXyNwl1kf7Q2e9PjlSyEyx+G4af6XZFIh9TmLhKq5np4choUz4ML/hdGt5+zRiRyaM9dJBSNNfDEVFj3vhfsx13nd0Ui+6RwF+nMroOnpZ/Ct/8OR032uyKRTincRfalZovX3XHbGpj8GIw63++KREKicBfZm8p18MhEqK2AK56CEaf7XJBI6BTuIh2pWAmPfAua6+CqOTCo0O+KRPaLwl2kvc2LvXlP4xK8E5T6H+l3RSL7TV0hRdpa/yE8fDEkpcO1ryrYpcdSuIvsUjzfO3iangfXvgL9DvG7IpEDpnAXAVj+nNePPXukt8fee5DfFYkcFIW7yCePwOxrYeCxcNWLkJbtd0UiB03hLrHtn/fAnB/CiDO8YXt79fG7IpGwUG8ZiU3Owdv/A+/8Hg6/GC69XzMoSVRRuEvsCQRg3s9hwb1w9DS46C6I10dBoov+oiW2tLbAizfDkplw/Pfh3P+GOLVOSvRRuEvsaKyF2ddA8WvwzVvg9FvAzO+qRLqEwl1iQ3UpPH45lH0WHLL3er8rEulSCneJfmXLYeblUF8JU5+EkeP9rkikyyncJbqteQueuhISU72zTvOP8rsikW6hI0kSvT55FGZe5p1tesMbCnaJKdpzl+jjHLz1W3j3Du/kpMsfhpTeflcl0q1C2nM3swlmttLMVpvZLXvZ5nIzW2Fmy83s8fCWKRKilkZ4droX7GOnwRVPK9glJnW6525m8cA9wDlACbDQzOY451a02aYA+E/gZOdcpZnldlXBIntVXwmzpsH69+HMX8CpP1FXR4lZoTTLjANWO+fWApjZLGAisKLNNjcA9zjnKgGcc+XhLlRknyrXwcxJ3vUl/4Axl/tdkYivQmmWGQhsbLNcElzX1khgpJl9YGYfmdmEjh7IzKabWZGZFVVUVBxYxSLtlSyC+8+G2nJv8C8Fu0hI4d7R/2tdu+UEoAA4HZgK3G9mXxtezzk3wzlX6JwrzMnJ2d9aRb7u8xfhoQu8ro7XzYdhp/hdkUhECCXcS4DBbZYHAZs72OYF51yzc+5LYCVe2It0nX/+DZ78LuSNhuvfgJyRflckEjFCCfeFQIGZDTezJGAKMKfdNs8DZwCYWTZeM83acBYqslugFV75D5j3nzDqArjqJUjX/wRF2ur0gKpzrsXMbgLmAfHAA8655WZ2O1DknJsTvG+8ma0AWoGfOue2dWXhEqOa671Zk1bOhRNuhPG/gbh4v6sSiTjmXPvm8+5RWFjoioqKfPnd0kMFWr2hBL54Gc77PRz/L35XJNLtzGyRc66ws+10hqr0DM55E2x88RJMULCLdEZjy0jP8NHfYMF9XlPMCd/zuxqRiKdwl8i3/HmYd6s31+n4//K7GpEeQeEukW3DR95YMYPHwSUzNCWeSIj0SZHItbUYnpjiDdk7dRYk9vK7IpEeQ+Eukam2HB67FCweps2G1L5+VyTSo6i3jESepp3w+GQv4K9+GfqO8LsikR5H4S6RJdAKz1wPpUtg8kwYdKzfFYn0SAp3iRzOwSs/884+Pf+PMOp8vysS6bHU5i6R48O7YeH9cNLNMO4Gv6sR6dEU7hIZPnsG5t8GR1wCZ//a72pEejyFu/hv3Qfw3PdgyEnwrXvVl10kDPQpEn9VrIRZUyFrGEyZCYkpflckEhUU7uKfmjJ47DKIT4YrnlZfdpEwUm8Z8UdjLTw+Ceq2en3Zs4b5XZFIVFG4S/drbYHZ18CWZd6wAgOP8bsikaijcJfu5RzM/TEUvwYX/hlGnut3RSJRSW3u0n1aW2DuT2HRQ3DKv0HhtX5XJBK1tOcu3aOxFp65Dla9Cif9EM68ze+KRKKawl26XnUpPH45lH0GF/wJjrvO74pEop7CXbpW2XKYeTnUV8LUJ2HkeL8rEokJCnfpOmvehKeugqQ0uPYVyD/K74pEYoYOqErX+OQRmDkJ+gyB699QsIt0M+25S3g5B2/+F7z3RzjkLJj0EKRk+l2VSMxRuEv4tDTC8z+Az2bDMVfBBf8L8Yl+VyUSkxTuEh5122HWFbDhQzj7V3Dyv4KZ31WJxCyFuxy8bWu89vWqErjsATjyUr8rEol5Cnc5OBs/hiemeG3tV82BISf4XZGIoN4ycjCWPwcPXQgpveH61xXsIhFE4S77zzl4/054+moYMBauex36HeJ3VSLSRkjhbmYTzGylma02s1v2sd1lZubMrDB8JUpEaW2Bl/8dXv+lN9/plS9AWj+/qxKRdjptczezeOAe4BygBFhoZnOccyvabZcB3Aws6IpCJQJUb4bZ18KGf3qjOp55m+Y7FYlQoXwyxwGrnXNrnXNNwCxgYgfb/Qb4A9AQxvokUqx+He47BUqXwiX3e90dFewiESuUT+dAYGOb5ZLgut3MbCww2Dn3Uhhrk0gQaPXOOH3sMkjPg+lvw5hJflclIp0IpStkR2eiuN13msUBfwau7vSBzKYD0wGGDBkSWoXin5ot8Mz1sO49GDsNzrsDklL9rkpEQhBKuJcAg9ssDwI2t1nOAI4E3jbvjMT+wBwzu9g5V9T2gZxzM4AZAIWFhQ6JXGvfhmdugMYa+Na9cPR3/K5IRPZDKOG+ECgws+HAJmAKsPuT7pyrArJ3LZvZ28BP2ge79BCBVnj3Dnj7d5A90jsxKfdwv6sSkf3Uabg751rM7CZgHhAPPOCcW25mtwNFzrk5XV2kdJPacnj2Bm+vfcwUb+Cv5HS/qxKRAxDS8APOubnA3HbrOpwE0zl3+sGXJd1u3fsw+zpo2AEX/wXGflcDf4n0YBpbJtYFAvD+n+Ct30LfETDtGeh/pN9VichBUrjHsp3b4LnpXh/2Iy+Fi+6C5Ay/qxKRMFC4x6oNH8HT10DdNrjwz3DsNWqGEYkiCvdYEwjAh3fDG7cH5zedr/lNRaKQwj3WvPYL+OgeGD3RO3Ca0tvvikSkCyjcY8k//+YF+7jpcN4f1AwjEsU08lOsWPECzPs5jLoQJvxOwS4S5RTusWDDAnh2OgwqhEvvh7h4vysSkS6mcI92W1d7c5xmDoCpsyCxl98ViUg3ULhHs9oKmHmp1wRzxWxIy+78Z0QkKuiAarRq2gmPXw41ZXD1S5rjVCTGKNyjUWuLN05M6RKY/JjX1i4iMUXhHm2cg1d+BqtegfP/CKMu8LsiEfGB2tyjzQd3QdH/wUk3w7gb/K5GRHyicI8my2bD67/0BgE7+9d+VyMiPlK4R4t178Pz34ehJ3vT4sXprRWJZUqAaFD+Bcz6DmQNhykzISHZ74pExGcK956uuhRmXgYJKTBtNvTK8rsiEYkA6i3TkzXWeH3Z67bDNXO9IXxFRFC491ytzfD01VC2HL7zFAw42u+KRCSCKNx7IufgpX/zpse7+C9QcLbfFYlIhFGbe0/07h2w+FE47WdwzJV+VyMiEUjh3tN8dB+89Vs46jtwxs/9rkZEIpSaZXoK5+DN38B7/+tNuHHRXZpwQ0T2SuHeE7S2wEv/6jXFHHMVXPAniNdbJyJ7p4SIdM31MPtaWDnXa2M/4+faYxeRTincI1l9JTwxFTZ85I3wqIHARCRECvdIVb0ZHrsUtq2GSQ/CEd/2uyIR6UEU7pGoYhU8dgnU7/CmxxvxTb8rEpEeRuEeaUoWeWPFxMV70+PpzFMROQDq5x5Jil+Hhy+ElEy4dp6CXUQOWEjhbmYTzGylma02s1s6uP/fzWyFmS01szfMbGj4S41yS5+CJyZ7E1lf+5omtBaRg9JpuJtZPHAPcB4wGphqZqPbbbYYKHTOjQFmA38Id6FR7Z/3wLM3wJAT4eq5kJHnd0Ui0sOFsuc+DljtnFvrnGsCZgET227gnHvLOVcXXPwIGBTeMqOUczD/Npj3cxg9EaY94zXJiIgcpFDCfSCwsc1ySXDd3lwHvNLRHWY23cyKzKyooqIi9CqjUWszPP8Db0Lr466Hyx7UDEoiEjah9Jbp6HRI1+GGZtOAQqDDvnvOuRnADIDCwsIOHyMmNNV5Y7EXz4MzboXTfqqzTkUkrEIJ9xJgcJvlQcDm9huZ2dnArcA3nXON4SkvCtVth8cnw6YiuPBOKLzG74pEJAqFEu4LgQIzGw5sAqYA32m7gZmNBf4OTHDOlYe9ymhRscqbFq96M0x6GEZf7HdFIhKlOg1351yLmd0EzAPigQecc8vN7HagyDk3B7gDSAeeNq95YYNzTsnV1uo34OlrICHJOzlp8Di/KxKRKBbSGarOubnA3HbrbmtzW/O87cuCGfDqLZB7OEx9QhNZi0iX0/ADXam12Qv1hffDYefDJf+A5HS/qxKRGKBw7yr1lV6PmLVvw8k/grN+6Y0XIyLSDRTuXWHbGu/AaeV6mPg3GHuF3xWJSIxRuIfb2nfgqSu9vfSrXoShJ/pdkYjEII0KGU5FD3jjsGfkww1vKthFxDfacw+H1hZ47Rew4F4oGA+X/p/GiBERXyncD1ZDlTeB9erX4YQbYfxvdOBURHyncD8Y29fC41Ng+xq46C449mq/KxIRARTuB27dB/DkNHAB+O5zMPw0vysSEdlNB1QPxOLH4JGJkNrPO3CqYBeRCKM99/3R0gTz/x8suA9GnAGTHoJeffyuSkTkaxTuoare7J1xunEBnPADOOc3EK+XT0Qik9IpFF++B7Ov8SbZuOwBOPJSvysSEdknhfu+OAcf3g2v/xr6HQJXvQS5o/yuSkSkUwr3vWmohhd+AJ+/6E1ePfEeSM7wu6qwaGkNkBDfM4+lO+eoa2olLVl/uiL7ok9IR8pWeN0cK9fB+N/CiTf22DlOWwOOL7ZUs2h9JUXrKlm0vpIt1Q2cOSqXywsHc/phOSRGeNBXNzTz4eqtvFu8lfeKK9i4vZ5DctI4tSCH00Zmc/zwfhEf9i2tATZW1rO6vJYtVfUcMzSL0fmZWA/9u5LIF9mfCD8sfRpevNnbS7/6JRh6kt8V7ZedjS0s3rCDovXbWbS+ksUbdlDb2AJAXmYyhUP7kpORzEtLS5m/oozs9GQuPWYgkwoHc2huZIw139IaYOmmKt5btZV3iytYsnEHrQFHenICJx7Sj0vGDmLJxh3MWriBhz5cR2K8cezQLE4tyOHUgmyOHNCbuDh/QrOhuZUvt+5kdXntVy5fbt1JU2vgK9sO7NOLc0bncc7oPMYN7xvxX7LSs5hzzpdfXFhY6IqKinz53R1qaYLXboWPZ8CQk2DSg5DR3++qOlVaVU/RukqK1m2naH0ln5dWE3DefzQOy8ugcFgWhUP7cuzQLAZl9dq9p9jcGuDtlRU8VbSRN78opzXgOHZoFpcXDuKCMQNI7+Y94ZLKOt4L7pm/X7yV6oYWzGDMwN6cNjKHUwtyGDukz1cCsKG5lUXrK3m3uIL3Vm1lRWk1AFmpiZx8aDanFeRwSkE2A/r0Cnu9NQ3Ne8K7opY15bUUl9eycXsdgeBHygyG9E3l0Jx0Ds1L965z08lOT+bDNVuZv6KM94q30tgSIDMlgTNG5XL24XmcflgOGSmJYa9ZooOZLXLOFXa6ncIdqNrkdXMs+RhOvAnO/hXER8aHq7Gllaq6ZnbUN7OjrpkddU2UVjWwaL3XxLJpRz0AvRLjOXpwHwqHZXHs0CzGDsmid6/QnkN5TQPPL97Ekws3sqZiJ6lJ8VzwjXwuP24whUOzuqTpYGdjCx+t3cZ7xd7e+dqKnQDk907h1IJsThuZw8mHZJOVlhTyY1bUNPLhmq28u8r7kiivaQTotAmnqSXAjvqm3a9z5c4mdtQ3U1XXTGVdU/C1bwq+/s1srW3c/dgAifHGiGwvuA/J9a4LctMZnp1GSuK+xxmqa2rh/WIv6N/4opztO5tIjDdOGNGP8aPzOHt0Hvm9w//lJD2Xwj1Ua9/xBv5qaYCJf4Ujvh32X9EacNQ2tFDd0Ex1QzM1DS3sqGumqr6JymBgVNXvCQ8vWLxQqWtq7fAxdzWxHDs0i8JhWRyen3nQ/613zvHJhh08XbSRFz/dzM6mVkZkpzGpcDCXHjOQ3MyU/Xq8uqYWSqsaKN3RQGlVvXe7qoE1FbUs3lBJc6sjJTGOE0b049SCHL45MptDctLD8mXinGNVWS3vFVfwbvFWPv5yGw3NARLjjdEDetPSGtj9ZblzL68xQEKc0Sc1kd69EumTmkRWaiJZqUkMz0nbvSc+pG9qWA5QtwYcn2yo5PUVZcxfUcbard4X3pEDMznn8P6cMzqPw/Mz1E4f4xTunXEOPrgT3rgd+hXA5McgZ2SnP1bd0MziDTuoqm+mut4Lai+wg7eD6/asb9nd5r03ifFGn9Qk+vRKDAZJEn1SE/csB0OlT3B9dnoyeZnJXfohr2tq4eWlpTxdVMLH67YTH2ecPjKHy48bzJmjcmlpdWyuqmdLVQObdwSvqxrY0ibEq+qbv/a4/dKSGNQ3lRNG9OW0ghyOHZrV6d5tOLRtwlm6sYrUpHjvNd/1Oqftef2zUpOCYZ5IenKCb2G6uryW+SvKmL9iC4s37sA5GJTVizNH5TI8O428zBRyM5LJy0whJyO5W15H8Z/CfV8aquC578PKl+GIS+Div+xz4uqmlgBvryznhSWbmf95GU0tXz0wlhBnZKQkkNkr0btO8a4zUhJ3395zn3d/72CI9ElNpFdifETvja2tqGX2ohJmLyqhvKaRpPi4rx0cBC+4+/dOIb93L/J7p5DfJ8W7Di7nZaYogA5QRU0jb37h7dG/v3orDc1ff/37pCaSl5FCbmYyuRkp5GV6wZ+XmUxOxq7rZJIT9B70ZAr3jgQC3rjrr/4H7NgA4/8Ljv9eh90cAwHHog2VPLd4E3OXlbKjrpl+aUlcOCafc4/oT25m8u7wTkmMi+hwDpeW1gDvFlfwwept9E1LYkCfFPpn9mJAHwV3d3LOUVnXTFl1A+U1jd51dQNl1d7tsppGKoL3tQS+/vnul5bE8Ow0Dg0eHzg0N52CvAwG9E6Jib/jnk7h3lbddlgyExbe7/VdzxwEl97f4TR4xWU1PL9kE88v3symHfWkJMZx7hH9+dbRAzmlIFvd1aTHCAQc2+uaguHfSHmN9wVQWlXPmvKdrK6oZfvOpt3bpybFe2HfpndPQV4Gg7N69diT3qJRqOEe3f3cNy+Bhf+AZbO9A6ZDToKzboNRF0HCnl4YZdUNvPjpZp5bvInlm6uJMzilIIefnDuS8aP7R/wJMiIdiYszstOTyU5P5ogBHW+zrbZxd3fO4rJa1lTU8uGabTy7eNPubZLi47w9/WDgj8zL4JRDs+mdGhk9yqRj0ZdaLY2w4gX4+B9e18bEVDhqChx3A/Q/cvdmNQ3NvPrZFl5YspkP1mzFOThqUG9uu3A0Fx6VT27G/vUMEemJ+qUn0y89meNH9PvK+uqGZta0OxFrWUkVc5eV4px3nOnkQ7O54Bv5nDM6b7+6rEr3iJ5mmaoSKHoQPnkYdlZA30Ng3A1w1FTo1YdttY2sKquluLyGBV9u5/UVZTS2BBjSN5VvHT2AiWMHckhOZJyhKRKpGppbWVFazbzlW5i7rJSN2+tJiDNOPKQf53/DOx7VV0HfpWKjzd05+PJd76zSlXMBaDpkPGuHTWVh/FEUl+9kVVkNxWW1bGvTttgvLYkLxuQz8eiBHDOkjw4iiRwA5xzLN1fz8rJS5i4rZf22OuLjjBNH9OO8b/Tn3CP6k52e7HeZUSe6w72hmvoi7wBpr6rV1MX35vVe5/KP+jNYtrP37s3SkxMoyEtnZG6Gd52Xwci8jC7vIy4Sa5xzrCitZu6yUuYu28KXW3cSZ3D88H6cPyafc4/IU1NnmERtuH88+08c8dnvSaOBJYERPNoynjcSTmZobt/d4b0ryPPVtUuk2znn+GJLDa8sK+XlZaWsqdiJGYwb1pfzv5HPhCP7k7efZzvLHmENdzObANwFxAP3O+d+1+7+ZOAR4FhgGzDZObduX495oOG++J0XaP7kcTYeegVZBcdTkJvBwD69fBsFUET2bVVZDS8vLeWVz0pZVVYLQEZyAjmZyeSkJ5ObmRK8TiY3wzvRKjfDO/u2T2qidtDaCVu4m1k8sAo4BygBFgJTnXMr2mzzA2CMc+57ZjYF+LaqwUWAAAAF4ElEQVRzbvK+Htf34QdEpNutLq/hzS/K2byjgYoar++9d93Y4ThKifFGTroX+DnBs29z0pPpl55E3zTv0i8tmb5p3hAdsdAfP5z93McBq51za4MPPAuYCKxos81E4FfB27OBv5qZOb/afEQkIh2am8GhuR3PaFbb2OIFffDs2l2hv+sLoKSyjsUbKr/SOaK93r0S6RcM/ay0pN2321+SE+JJiDcS4+JIiLev3E6MjyMhzoiPsx79v4ZQwn0gsLHNcglw/N62cc61mFkV0A/YGo4iRST6pScnkJ6cwPDstH1u19waoLKuie07v3rZVttEZV0T23Y2sb22iQ3b6liycQeVO5s6HIYhFInxRkKcF/bel0AciW1C32zP6CWGtbnNnm12/7Nn/c1nFXDxUXs5syxMQgn3jr662r9SoWyDmU0HpgMMGTIkhF8tIvJVifFxwTb50A7KOueorm9h287G4JdCM00tAVoCAZpbHS2tAZoD3nVLq6M54F23Xd/c6mgNuN0/4xw43O6Uc8Hfs+d2x+t3bd8nxLkWDkYo4V4CDG6zPAjYvJdtSswsAegNbG//QM65GcAM8NrcD6RgEZH9YWb0Tk2MueESQjn6sBAoMLPhZpYETAHmtNtmDnBV8PZlwJtqbxcR8U+ne+7BNvSbgHl4XSEfcM4tN7PbgSLn3Bzg/4BHzWw13h77lK4sWkRE9i2kgcOcc3OBue3W3dbmdgMwKbyliYjIgYr+TqEiIjFI4S4iEoUU7iIiUUjhLiIShRTuIiJRyLchf82sAljvyy/3XzaxPTRDrD9/0Gug53/gz3+ocy6ns418C/dYZmZFoYzqFq1i/fmDXgM9/65//mqWERGJQgp3EZEopHD3xwy/C/BZrD9/0Gug59/F1OYuIhKFtOcuIhKFFO5dzMwGm9lbZva5mS03sx8F1/c1s/lmVhy8zvK71q5kZvFmttjMXgouDzezBcHn/2RwOOmoZGZ9zGy2mX0R/Ds4MZbefzP7t+Df/mdm9oSZpUTz+29mD5hZuZl91mZdh++3ee42s9VmttTMjglXHQr3rtcC/Ng5dzhwAnCjmY0GbgHecM4VAG8El6PZj4DP2yz/Hvhz8PlXAtf5UlX3uAt41Tk3CjgK73WIifffzAYCNwOFzrkj8YYNn0J0v/8PARPardvb+30eUBC8TAfuDVsVzjlduvECvACcA6wE8oPr8oGVftfWhc95UPAP+kzgJbxpGbcCCcH7TwTm+V1nFz33TOBLgse32qyPifefPfMr98UbYvwl4Nxof/+BYcBnnb3fwN+BqR1td7AX7bl3IzMbBowFFgB5zrlSgOB1rn+Vdbk7gZ8BgeByP2CHc64luFyCFwLRaARQATwYbJa638zSiJH33zm3CfgjsAEoBaqARcTO+7/L3t7vXV9+u4TttVC4dxMzSweeAf7VOVftdz3dxcwuBMqdc4varu5g02jttpUAHAPc65wbC+wkSptgOhJsW54IDAcGAGl4TRHtRev735ku+ywo3LuBmSXiBftM59yzwdVlZpYfvD8fKPervi52MnCxma0DZuE1zdwJ9AlOpg4dT7oeLUqAEufcguDybLywj5X3/2zgS+dchXOuGXgWOInYef932dv7XQIMbrNd2F4LhXsXMzPDm2P2c+fcn9rc1XZS8avw2uKjjnPuP51zg5xzw/AOpL3pnLsCeAtvMnWI7ue/BdhoZocFV50FrCBG3n+85pgTzCw1+FnY9fxj4v1vY2/v9xzgymCvmROAql3NNwdLJzF1MTM7BXgPWMaeNuef47W7PwUMwfsATHLObfelyG5iZqcDP3HOXWhmI/D25PsCi4FpzrlGP+vrKmZ2NHA/kASsBa7B27GKifffzH4NTMbrObYYuB6vXTkq338zewI4HW/kxzLgl8DzdPB+B7/w/orXu6YOuMY5VxSWOhTuIiLRR80yIiJRSOEuIhKFFO4iIlFI4S4iEoUU7iIiUUjhLiIShRTuIiJRSOEuIhKF/j/IO9/KATmmPQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Building PDF and CDF plots\n",
    "counts, bin_edges = np.histogram(rawdata['Ins_Age']*100, bins = 20, density = True)\n",
    "pdf = counts/sum(counts)\n",
    "print(pdf)\n",
    "print (bin_edges)\n",
    "\n",
    "cdf = np.cumsum(pdf)\n",
    "plt.plot(bin_edges[1:],pdf)\n",
    "plt.plot(bin_edges[1:],cdf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we can see that the most of the applicant are around the age of 25, 45 and 60. Almost 98% of the applicant are below the age of 72 years."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's try to figure out the relationship between the Age and the risk factors(response) and see how different age groups have different risk factors associated with their applications."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x432 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.jointplot(x='Ins_Age',y='Response', data=rawdata, kind='kde')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we can see that this plot tells lot of details. We can see from the plot that it is less risky for the company to give insurance to the applicants who are ono the younger side of the range. Lot of applicants have risk factor of 8 specially in the age range of 0 to 40 which can be seen by the **GREY** bars and line. We can also see that many applicants having a risk factor of 1 who are above the age of 60. It is represented by the **BLUE** bars and line showcasing the probability distribution function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\mites\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6462: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n",
      "  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n",
      "C:\\Users\\mites\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6462: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n",
      "  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n",
      "C:\\Users\\mites\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6462: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n",
      "  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n",
      "C:\\Users\\mites\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6462: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n",
      "  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n",
      "C:\\Users\\mites\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6462: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n",
      "  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n",
      "C:\\Users\\mites\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6462: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n",
      "  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n",
      "C:\\Users\\mites\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6462: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n",
      "  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n",
      "C:\\Users\\mites\\Anaconda3\\lib\\site-packages\\matplotlib\\axes\\_axes.py:6462: UserWarning: The 'normed' kwarg is deprecated, and has been replaced by the 'density' kwarg.\n",
      "  warnings.warn(\"The 'normed' kwarg is deprecated, and has been \"\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x20b03a88a90>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 720x720 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 272.25x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,10))\n",
    "sns.FacetGrid(rawdata, hue = 'Response').map(sns.distplot,'Ins_Age').add_legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is a very interesting plot which shows the relation of a person's BMI with the risk factor. It showcases that there are lot of people in the BMI range of 20 and 60. Lot of people with BMI between 20 and 50 have risk factor of 8 and the risk factor gradually decreases as we can see from the change of colors in the plot. As the BMI goes on incresing we can see that risk factor decreses to 2 and 1, which means that it is very risky for the insurance company to give those applicants insurance."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After plotting several graphs we can see that there are some data points which seems bit off than the other regular data points. It can be outliers. One of the ways to detect outliers is to plot boxplots of columns to figure them out."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "___________________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Finding Anamolies"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's check if there are anamolies in Weight, Height, BMI and age features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20b0468ad30>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,10))\n",
    "rawdata.boxplot(column=['Ins_Age','Ht','Wt','BMI'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that there are ouliers in the dataset and need to be removed from the dataset for better model predictions of risk factors. One of the ways to get rid of outliers is to use Z-score to find data points within 3 standard deviation and discard the rest."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before jumping in directly to remove outliers, we need to make sure to deal with categorical data that can be useful for predictions.\n",
    "Checking for categorical features. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "cols = rawdata.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "num_cols = rawdata._get_numeric_data().columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Product_Info_2']"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(set(cols) - set(num_cols))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is just one column in the dataset which is categorical and it needs to be converted to numerical either by one hot encoding or using label encoding. Since there are lot of features and doing one-hot encoding is only going to add more complexity for the model during prediction."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "def label_encoding(data):\n",
    "    \n",
    "    from sklearn.preprocessing import LabelEncoder\n",
    "    data = LabelEncoder().fit_transform(data)\n",
    "    \n",
    "    return data\n",
    "\n",
    "rawdata['Product_Info_2'] = label_encoding(rawdata['Product_Info_2'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Removing ouliers outside 3 standard deviations"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Getting rid of ouliers witht the help of Z-score where only datapoints within 3rd quartile are considered."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy import stats\n",
    "rawdata = rawdata[(np.abs(stats.zscore(rawdata)) < 3).all(axis=1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20b5521aba8>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlYAAAJDCAYAAADJvlo3AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAHyJJREFUeJzt3X2QXXd93/HP15KwPNgxxQ5KignLtE6zqgiQUYCAa6/qhDFxit02abyZPNDRREOnVpvayVgZdSAh1VRKSjNToBA7onEYIkLSmaJBxmYm3TutCKSY8tDYO0xcIoJwebTrImKMJX79Y1eelZCtK+1POntXr9fMzu699+w93909Wr33nHPvrdZaAABYvouGHgAAYLUQVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADpZO9SKr7zyyjY1NTXU6lesb3zjG3n2s5899BhMANsKZ8L2wrhsK6f28Y9//Kutte8+3XKDhdXU1FTuv//+oVa/Yo1Go8zMzAw9BhPAtsKZsL0wLtvKqVXV58ZZzqFAAIBOhBUAQCfCCgCgE2EFANDJacOqqt5VVV+uqj9/mturqv5DVT1UVZ+uqh/qPyYAwMo3zh6r30tywzPc/tokVy++bUvyjuWPBQAweU4bVq21/5bkkWdY5KYkv98WfDTJc6rqe3sNCAAwKXqcY/X8JJ9fcvnw4nUAABeUHk8QWqe4rp1ywaptWThcmA0bNmQ0GnVY/epy5MgR3xfGYlvhTNheGJdtZXl6hNXhJC9YcvmqJA+fasHW2p1J7kySzZs3N8/s+p084y3jsq1wJmwvjMu2sjw9DgXuT/Lzi48OfGWSx1pr/6fD/QIATJTT7rGqqn1JZpJcWVWHk7wpybokaa29M8k9SX48yUNJ/jrJPz1XwwIArGSnDavW2uxpbm9J/nm3iQAAJpRnXgcA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2G1Quzbty+bNm3K9ddfn02bNmXfvn1DjwQAnKEerxXIMu3bty87d+7M3r17c+zYsaxZsyZbt25NkszOPuPzswIAK4g9VivArl27snfv3mzZsiVr167Nli1bsnfv3uzatWvo0QCAMyCsVoD5+flcc801J1x3zTXXZH5+fqCJAICzIaxWgOnp6Rw8ePCE6w4ePJjp6emBJgIAzoawWgF27tyZrVu3Zm5uLkePHs3c3Fy2bt2anTt3Dj0aAHAGnLy+Ahw/QX379u2Zn5/P9PR0du3a5cR1AJgwwmqFmJ2dzezsbEajUWZmZoYeBwA4Cw4FAgB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsFoh9u3bl02bNuX666/Ppk2bsm/fvqFHAgDO0NqhB2Ahqnbu3Jm9e/fm2LFjWbNmTbZu3ZokmZ2dHXg6AGBc9litALt27crevXuzZcuWrF27Nlu2bMnevXuza9euoUfjHKuqs37bsmXLsj6/qob+8gFWHWG1AszPz+eaa6454bprrrkm8/PzA03E+dJaO+u3F97xgWV9fmtt6C8fYNURVivA9PR0Dh48eMJ1Bw8ezPT09EATAQBnQ1itADt37szWrVszNzeXo0ePZm5uLlu3bs3OnTuHHg0AOANOXl8Bjp+gvn379szPz2d6ejq7du1y4joATBhhtULMzs5mdnY2o9EoMzMzQ48DAJwFhwIBADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnXhJm86qaugR0lobegQAuCDZY9VZa21Zby+84wPLvg8AYBjCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANDJ2qEHAACGd8UVV+SRRx556vJzn/vcfO1rXxtwoslkjxUAXOBOjqokeeSRR3LFFVcMNNHkElYAcIE7HlVVdcL7k2OL0xNWAECSZM2aNSe858wJKwAgSbJnz5588IMfzJ49e4YeZWI5eR0ASJLcfvvtQ48w8eyxAgDoRFgBAHQirACAJE5e70FYAQBJkmPHjp3wnjMnrACAJMn69evztre9LevXrx96lInlUYEAQJLkm9/8Zm699dahx5ho9lgBAHQirAAAOhFWAACdCCsAgE6EFQCQJJmamsq73/3uTE1NDT3KxBorrKrqhqr6TFU9VFU7TnH791XVXFV9oqo+XVU/3n9UAOBcOnToUH7u534uhw4dGnqUiXXasKqqNUnenuS1STYmma2qjSct9q+TvK+19rIktyT5j70HBQBY6cbZY/XyJA+11j7bWvtWkvcmuemkZVqS71r8+PIkD/cbEQBgMozzBKHPT/L5JZcPJ3nFScv8WpIPVdX2JM9O8qNdpgMAmCDjhFWd4rp20uXZJL/XWntLVf1IkndX1abW2rdPuKOqbUm2JcmGDRsyGo3OYuTVz/eFcdlWGNeRI0dsL5wV282ZGSesDid5wZLLV+U7D/VtTXJDkrTWPlJV65NcmeTLSxdqrd2Z5M4k2bx5c5uZmTm7qVezew/E94Wx2FY4A6PRyPbCWbHdnJlxzrH6WJKrq+pFVfWsLJycvv+kZf4qyfVJUlXTSdYn+UrPQQEAVrrThlVr7WiSW5Pcl2Q+C4/+e6Cq3lxVr1tc7PYkv1hVn0qyL8nrW2snHy4EAFjVxjkUmNbaPUnuOem6Ny75+MEkr+47GgDAZPHM6wAAnQgrAIBOhBUAQCfCCgCgE2EFACRJquqE95w5YQUAJEmOP1OSZ0w6e8IKAKATYQUA0ImwAgDoRFgBAHQirAAAOhFWAACdCCsAgE7WDj0ATLKX/PqH8tjjTw62/qkdBwZb9+WXrMun3vSawdYPsBIJK1iGxx5/Mod23zjIukejUWZmZgZZdzJs1AGsVA4FAgB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADpZO/QAMMkum96RF9+9Y7gB7h5u1ZdNJ8mNww0AsAIJK1iGr8/vzqHdw8TFaDTKzMzMIOtOkqkdBwZbN/DMqmrQ+2qtdVv/pBFWALDKnGnYPFM8XciRdDacYwUAF7iniydRdeaEFQCQ1lpaa3nhHR946mPOnLACAOhEWAEAdCKsAAA6EVYAAJ0IKwCAToQVAEAnwgoAoBNhBQDQibACAOhEWAEAdCKsAAA6EVYAAJ0IKwCAToQVAEAnwgoAoBNhBQDQibACAOhEWAEAdCKsAAA6EVYAAJ0IKwCAToQVAEAnwgoAoBNhBQDQibACAOhEWAEAdCKsAAA6EVYAAJ0IKwCAToQVAEAnwgoAoBNhBQDQydqhB4BJN7XjwHArv3e4dV9+ybrB1g2wUgkrWIZDu28cbN1TOw4Mun4AvpNDgQAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANDJWGFVVTdU1Weq6qGq2vE0y/yTqnqwqh6oqj/oOyYAwMq39nQLVNWaJG9P8mNJDif5WFXtb609uGSZq5P8apJXt9YerarnnauBAQBWqnH2WL08yUOttc+21r6V5L1JbjppmV9M8vbW2qNJ0lr7ct8xAQBWvnHC6vlJPr/k8uHF65b6/iTfX1UfrqqPVtUNvQYEAJgUpz0UmKROcV07xf1cnWQmyVVJ/ntVbWqt/d8T7qhqW5JtSbJhw4aMRqMznfeC4PvCuGwrjOvIkSO2F8ZmWzl744TV4SQvWHL5qiQPn2KZj7bWnkzyl1X1mSyE1seWLtRauzPJnUmyefPmNjMzc5Zjr2L3HojvC2OxrXAGRqOR7YXx+N2yLOMcCvxYkqur6kVV9awktyTZf9Iy/yXJliSpqiuzcGjwsz0HBQBY6U4bVq21o0luTXJfkvkk72utPVBVb66q1y0udl+Sr1XVg0nmkvxKa+1r52poAICVaJxDgWmt3ZPknpOue+OSj1uS2xbfAAAuSJ55HQCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBXAKrZ9+/asX78+W7Zsyfr167N9+/ahR4JVbe3QAwBwbmzfvj3vfOc7s2fPnmzcuDEPPvhg7rjjjiTJW9/61oGng9VJWAGsUnfddVeuvPLK3H777U9d9z3f8z256667hBWcIw4FAqxSTzzxRL74xS/m0ksvTZJceuml+eIXv5gnnnhi4Mlg9bLHCmAVW7t2bfbv359jx45lzZo1ec1rXpOjR48OPRasWsIKYBU7evRoZmdn8+UvfznPe97zRBWcY8IKYJX70pe+dMJ74NxxjhUAQCfCCgCgE2EFANCJsAJY5ZY+3QJwbgkrgFVu3bp1ueiii7Ju3bqhR4FVz6MCAVa5Rx999IT3wLljjxUAQCfCCgCgE2EFANCJsAIA6ERYAaxyxx8N6FGBcO4JK4BV7sknnzzhPXDuCCsAgE6EFcAqd9FFF53wHjh3/CsDWOW+/e1vn/AeOHeEFQBAJ8IKAKATYQUA0ImwAgDoRFgBAHQirABWuUsvvfSE98C5s3boAQA4t44cOXLCe1a+l/z6h/LY48M9U/7UjgODrfvyS9blU296zWDrXy5hBQArzGOPP5lDu28cZN2j0SgzMzODrDsZNup6cCgQAKATYQUA0ImwAgDoRFgBAHQirAAAOvGoQBhQVS3v8/csb/2tteXdAQAnsMcKBtRaO+u3ubm5ZX2+qALoT1gBAHTiUCDABFjuYeMe92cvJ5yesAKYAGcTNc8UTyIJzg2HAgFWqaeLJ1EF546wAljFjj9Q4YV3fMCDFuA8cCjwJEO/ongy3AtQTvorigPA0ITVSYZ8RfFk2FcVn/RXFAeAoTkUCADQibACAOhEWAEAdCKsAAA6EVYAAJ0IKwCAToQVAEAnwgoAoBNhBQDQibACAOhEWAEAdCKsAAA6EVYAAJ0IKwCAToQVAEAnwgoAoBNhBQDQibACAOhEWAEAdCKsAAA6EVYAAJ0IKwCAToQVAEAna4ceAAA40WXTO/Liu3cMN8Ddw636sukkuXG4AZZJWAHACvP1+d05tHuYuBiNRpmZmRlk3UkytePAYOvuwaFAAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOxgqrqrqhqj5TVQ9V1dM+x35V/WRVtara3G9EAIDJcNqwqqo1Sd6e5LVJNiaZraqNp1jusiT/Ismf9R4SAGASjLPH6uVJHmqtfba19q0k701y0ymW+40kv5nkmx3nAwCYGOO8CPPzk3x+yeXDSV6xdIGqelmSF7TWPlBVv/x0d1RV25JsS5INGzZkNBqd8cDnw5BzHTlyZND1r9SfCd9p6G2FyWN7mSxD/bxWwu+Wode/HOOEVZ3iuvbUjVUXJfntJK8/3R211u5McmeSbN68uQ356tlP694Dg76q96CvKj7w186ZGfoV6Jkw/n1PlgF/XoP/bpnwbXWcQ4GHk7xgyeWrkjy85PJlSTYlGVXVoSSvTLLfCewAwIVmnLD6WJKrq+pFVfWsJLck2X/8xtbaY621K1trU621qSQfTfK61tr952RiAIAV6rRh1Vo7muTWJPclmU/yvtbaA1X15qp63bkeEABgUoxzjlVaa/ckueek6974NMvOLH8sAIDJ45nXAQA6EVYAAJ0IKwCAToQVAEAnwgoAoBNhBQDQibACAOhEWAEAdCKsAAA6EVYAAJ2M9ZI2AMD5NbXjwHArv3e4dV9+ybrB1t2DsAKAFebQ7hsHW/fUjgODrn/SORQIANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0snboAVaay6Z35MV37xh2iLuHWe1l00ly4zArB4BVQFid5Ovzu3No93BxMRqNMjMzM8i6p3YcGGS9ALBaCCuA8+Alv/6hPPb4k4POMNQfT5dfsi6fetNrBlk3nG/CCuA8eOzxJ+0NhwuAk9cBADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBO1g49AMCF4LLpHXnx3TuGHeLuYVZ72XSS3DjMyuE8E1YA58HX53fn0O7h4mI0GmVmZmaQdU/tODDIemEIDgUCAHQirAAAOhFWAACdCCsAgE6EFQBAJ8IKAKATYQUA0ImwAgDoRFgBAHQirAAAOhFWAACdCCsAgE6EFQBAJ8IKAKATYQUA0ImwAgDoRFgBAHQirAAAOhFWAACdCCsAgE6EFQBAJ8IKAKATYQUA0ImwAgDoZO3QAwBcKKZ2HBh2gHuHWf/ll6wbZL0wBGEFcB4c2n3joOuf2nFg8BngQuBQIABAJ8IKAKATYQUA0MlYYVVVN1TVZ6rqoaracYrbb6uqB6vq01X1J1X1wv6jAgCsbKcNq6pak+TtSV6bZGOS2araeNJin0iyubX2g0n+OMlv9h4UAGClG2eP1cuTPNRa+2xr7VtJ3pvkpqULtNbmWmt/vXjxo0mu6jsmAMDKN87TLTw/yeeXXD6c5BXPsPzWJB881Q1VtS3JtiTZsGFDRqPReFOeZ0POdeTIkUHXv1J/JnynobcVJo/thXHZVs7eOGFVp7iunXLBqp9NsjnJdae6vbV2Z5I7k2Tz5s1tZmZmvCnPp3sPZMi5RqPRcOsf+GvnzAy6rTB5/PtmXLaVZRknrA4necGSy1clefjkharqR5PsTHJda+2JPuMBAEyOccLqY0murqoXJflCkluS/MzSBarqZUl+J8kNrbUvd5/yPPOyEwDA2ThtWLXWjlbVrUnuS7Imybtaaw9U1ZuT3N9a25/kt5JcmuSPqipJ/qq19rpzOPc5M/RLPnjZCQCYXGO9VmBr7Z4k95x03RuXfPyjnecCAJg4nnkdAKATYQUA0ImwAgDoRFgBAHQirAAAOhnrUYEAwORYfOqjs//8Pctbf2unfIGWC4I9VgCwyrTWzvptbm5uWZ9/IUdVIqwAALoRVgAAnQgrACDbt2/P+vXrs2XLlqxfvz7bt28feqSJ5OR1ALjAbd++Pe985zuzZ8+ebNy4MQ8++GDuuOOOJMlb3/rWgaebLPZYAcAF7q677sqePXty2223Zf369bntttuyZ8+e3HXXXUOPNnGEFQBc4J544om84Q1vOOG6N7zhDXniiScGmmhyORQIABe4iy++ONu2bcsnP/nJzM/PZ3p6Oi996Utz8cUXDz3axLHHCgAucNddd13e85735Nprr8373//+XHvttXnPe96T6667bujRJo49VgBwgfvCF76Qm2++Oe9617vyjne8IxdffHFuvvnm/MVf/MXQo00cYQUAF7j5+fl84hOfyLp16zIajTIzM5Mnn3wy69evH3q0ieNQIABc4Kanp3Pw4METrjt48GCmp6cHmmhyCSsAuMDt3LkzW7duzdzcXI4ePZq5ubls3bo1O3fuHHq0ieNQIABc4GZnZ5MsPFHo8UcF7tq166nrGZ+wAgAyOzub2dnZp86x4uw4FAgA0ImwAgDoRFgBAHQirAAAOhFWAACdCCsAgE6EFQBAJ8IKAKATYQUA0ImwAgDoRFgBAHQirAAAOhFWAACdCCsAgE6EFQBAJ8IKAKATYQUA0ImwAgDoRFgBAHQirAAAOhFWAACdCCsAgE6EFQBAJ8IKAKATYQUA0ImwAgDoRFgBAHQirAAAOhFWAACdCCsAgE6EFQBAJ8IKAKATYQUA0ImwAgDoRFgBAHQirAAAOhFWAACdCCsAgE6EFQBAJ8IKAKATYQUA0ImwAgDoZO3QAwBwelW1/PvYs7zPb60tewZY7eyxApgArbVlvc3NzS37PoDTE1YAAJ0IKwCAToQVAEAnwgoAoBNhBQDQibACAOhEWAEAdCKsAAA6EVYAAJ0IKwCAToQVAEAnwgoAoBNhBQDQibACAOhEWAEAdCKsAAA6EVYAAJ0IKwCAToQVAEAnwgoAoBNhBQDQibACAOhEWAEAdCKsAAA6EVYAAJ0IKwCAToQVAEAnwgoAoJOxwqqqbqiqz1TVQ1W14xS3X1xVf7h4+59V1VTvQQEAVrrThlVVrUny9iSvTbIxyWxVbTxpsa1JHm2t/e0kv51kT+9BAQBWunH2WL08yUOttc+21r6V5L1JbjppmZuS3L348R8nub6qqt+YAAAr39oxlnl+ks8vuXw4ySuebpnW2tGqeizJFUm+unShqtqWZFuSbNiwIaPR6OymXsG2bNmy7PuoZe7vm5ubW/YMrHxHjhxZlf+GODdsL4zLtrI844TVqfY8tbNYJq21O5PcmSSbN29uMzMzY6x+srT2HV/2GRmNRlmN3xf6s61wJmwvjMu2sjzjHAo8nOQFSy5fleThp1umqtYmuTzJIz0GBACYFOOE1ceSXF1VL6qqZyW5Jcn+k5bZn+QXFj/+yST/tS131w0AwIQ57aHAxXOmbk1yX5I1Sd7VWnugqt6c5P7W2v4ke5O8u6oeysKeqlvO5dAAACvROOdYpbV2T5J7TrrujUs+/maSn+o7GgDAZPHM6wAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfCCgCgE2EFANCJsAIA6ERYAQB0IqwAADoRVgAAnQgrAIBOhBUAQCfVWhtmxVVfSfK5QVa+sl2Z5KtDD8FEsK1wJmwvjMu2cmovbK199+kWGiysOLWqur+1tnnoOVj5bCucCdsL47KtLI9DgQAAnQgrAIBOhNXKc+fQAzAxbCucCdsL47KtLINzrAAAOrHHCgCgE2EFE6Kqjpx0+fVV9bbFj2+uqo3DTMZKUVW/XVW/tOTyfVX1u0suv6WqbquqnxlmQlaCqjpWVZ+sqk9V1f+sqlctXj9VVa2qfmPJsldW1ZNLftf8WlX98lCzTwJh1cHJ/+F1us9PVdW+3vfLqnVzEmHFnyY5/p/kRVl4PqK/u+T2VyVZl0RYXdgeb629tLX2kiS/muTfLrnts0l+Ysnln0rywPkcbtIJqxWoqqaz8LO5tqqePfQ8rGyLf22+LslvLf4V+reGnonBfDiLYZWFoPrzJF+vqr9RVRcnmU7y00n+3uK28q8GmpOV47uSPLrk8uNJ5qvq+PNY/XSS9533qSbY2qEHWE2qaibJr2XhGWs3Jfl4kp9trbWq2p2F//yOJvlQa+2ZdqX+TJJ3Z+GX4OuS7Fu8/x9OsjfJN5IcTPLa1tqmqlqTZHeSmSQXJ3l7a+13en99DO6SqvrkksvPTbK/tfanVbU/yQdaa3880GysAK21h6vqaFV9XxYC6yNJnp/kR5I8luTTSd6Y5Jdbaz/x9PfEKnf8d8n6JN+b5O+fdPt7k9xSVV9McizJw0n+5vkdcXIJq/5eloW/FB/Owl+Pr66qB5P8wyQ/sBhZzznNffx0kh9L8neS3JrFsEryn5JsW/yPdPeS5bcmeay19sOLf5V+uKo+1Fr7y35fFivA4621lx6/UFWvT+LZkTnZ8b1Wr0ry77MQVq/KQlj96YBzsXI89bukqn4kye9X1aYlt9+b5DeSfCnJHw4w30RzKLC//9FaO9xa+3aSTyaZSvL/knwzye9W1T9K8tdP98mLe6W+0lr7XJI/SfJDi7vxn5Pkstba8V+Mf7Dk016T5OcX/wL5syRXJLm689cFTIbj51m9OAuHAj+ahT1Wr8pCdMFTWmsfycK5eN+95LpvZeGIy+1J/vNAo00sYdXfE0s+PpZkbWvtaJKXZ2EDvTkLfw08ndkkP1BVh5L87ywc//7HSeoZPqeSbF88GfGlrbUXtdY+tIyvgcnz9SSXDT0EK8KHs3Dy8SOttWOttUeSPCcLcfWR2FZYoqp+IMmaJF876aa3JLmjtXby9ZyGsDoPqurSJJe31u5J8ktJXvo0y12UhUdg/GBrbaq1NpXkpiSzrbVHs3AS6isXF79lyafel+SfVdW6xfv5fie9X3Dem+RXquoTTl6/4P2vLOyB+OhJ1z3WWvtqFs6zOrr4yGMnr1+YLll88MIns3Co7xdaa8eWLtBae6C1dvcw4002z7zeQVUdaa1dunjy+lMnhS4+78f9WQif92fhRMFK8u9OtcEufv7u1torl1y3JsnhJD+U5PuS3JWFk9dHSa5trb16Mcj+TZJ/sHj/X0lyc2vtsXPyBQMApySsJkhVXdpaO7L48Y4k39ta+5cDjwUALPKowMlyY1X9ahZ+bp9L8vphxwEAlrLHaiBVtTML51Mt9UettV1DzAMALJ+wAgDoxKMCAQA6EVYAAJ0IKwCAToQVAEAnwgoAoJP/DztVbRnRHH9QAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,10))\n",
    "rawdata.boxplot(column=['Ins_Age','Ht','Wt','BMI'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we can see that most of the ouliers have been removed and only the ones which were within the 3rd standard deviation are kept for further processing."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's plot a corellation matrix to see the degree of corellation between age, weight, height, BMI and response(risk factor)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20b044b1828>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def make_corr_matrix(data):\n",
    "    \n",
    "    corr = data.corr()\n",
    "    ax = sns.heatmap(corr, vmin=-1, vmax=1, center=0,\n",
    "                    cmap = sns.diverging_palette(20,220, n=200),\n",
    "                    square = True)\n",
    "    ax.set_xticklabels(ax.get_xticklabels(), rotation=45,\n",
    "                      horizontalalignment = 'right');\n",
    "    \n",
    "    return ax\n",
    "\n",
    "make_corr_matrix(rawdata[['Ins_Age','Ht','Wt','BMI','Response']])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the above corellation matrix we can see there is corellation between BMI, Age, Weight, Height and the response (risk factor) for the insurance applicants.\n",
    "\n",
    "There is a good negative corellation between the features and the risk factors. We can confirm this from the dataset as BMI increases we have the less respnse value which relates to more risk factor. So this corellation makes sense because features will be inversely related to risk factor."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "________________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Processing data for machine learning models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = rawdata.iloc[:,:-1] #Selecting all columns except the last target column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y = rawdata['Response']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(19501, 115)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finding the number of data points for each classification class. For a good model it is advised to have equal number of data points for each class so that the model doesn't overfit on the biggest class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,5))\n",
    "rawdata.Response.value_counts().plot(kind='bar');\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see from the resposnses that the dataset is imbalanced and if we try to build a model to predict the responses the model would be baised towards predicting classes that have lot of data points to train on.\n",
    "\n",
    "In order to deal with the imbalance we have to do data augmentation on the class with least data points or on class with max data points."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Deciding on whether to undersample or oversample the class depends on the the problem that's being solved. We can see that there are around 18000 datapoints for response value of 8 while there is only 1500 data points for response 3 and 4. When compared with the other responses it makes sense to over sample the minority classes. While oversampling the minority class we need to take care that the over sampling shouldn't lead to model overfitting on those classes. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\mites\\Anaconda3\\lib\\site-packages\\h5py\\__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
      "  from ._conv import register_converters as _register_converters\n",
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "def resampling_data(X,Y):\n",
    "    from imblearn.over_sampling import SMOTE\n",
    "    sm = SMOTE(sampling_strategy = 'not majority')\n",
    "    X,Y = sm.fit_resample(X,Y)\n",
    "    \n",
    "    return X,Y\n",
    "\n",
    "X_resampled, Y_resampled = resampling_data(X,Y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[   1 9293]\n",
      " [   2 9293]\n",
      " [   3 9293]\n",
      " [   4 9293]\n",
      " [   5 9293]\n",
      " [   6 9293]\n",
      " [   7 9293]\n",
      " [   8 9293]]\n"
     ]
    }
   ],
   "source": [
    "unique, counts = np.unique(Y_resampled, return_counts=True)\n",
    "print(np.asarray((unique, counts)).T)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After over sampling we can see that all the classes now have same number of datapaoints to works with so that whatever model is built to predict the risk factors doesn't overfit"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Splitting the datapoints for training and testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, Y_train, Y_test = train_test_split(X_resampled,Y_resampled, test_size = 0.20, random_state = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(14869, 115)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[3.80924982e+04, 1.00000000e+00, 1.62023374e+01, ...,\n",
       "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00],\n",
       "       [4.90393016e+04, 1.00000000e+00, 1.30071509e+01, ...,\n",
       "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00],\n",
       "       [5.74874527e+04, 1.00000000e+00, 7.00000000e+00, ...,\n",
       "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00],\n",
       "       ...,\n",
       "       [5.80689523e+04, 1.00000000e+00, 1.41316554e+01, ...,\n",
       "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00],\n",
       "       [3.89012898e+04, 1.00000000e+00, 1.56583496e+01, ...,\n",
       "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00],\n",
       "       [2.07980000e+04, 1.00000000e+00, 7.00000000e+00, ...,\n",
       "        0.00000000e+00, 0.00000000e+00, 0.00000000e+00]])"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Standardizing the data and applying PCA"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since we have values that are of various ranges across the features, we need to scale them so that they are standardized within a particular range and it would help the PCA while dimension reduction. It helps the PCA select the components with maximum variance."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "def standardizing_data(X_train,X_test):\n",
    "    from sklearn.preprocessing import StandardScaler\n",
    "    feature_scaler = StandardScaler()\n",
    "    X_train = feature_scaler.fit_transform(X_train)\n",
    "    X_test = feature_scaler.fit_transform(X_test)\n",
    "    \n",
    "    return X_train, X_test\n",
    "\n",
    "X_train, X_test = standardizing_data(X_train, X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After standardization of the data points, its time to apply PCA for reducing the dimensions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here in this case we want the data points to have variance so that while reducing diension through PCA, it will make the datpoints easily classifiable by the machine learning models"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The other task is to find the optimum number of components to select. We want to maximize the variance which can be determined by looking at the cumulative explained variance ratio as a function of the number of components."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def test_PCA(X_train):\n",
    "    from sklearn.decomposition import PCA as RandomizedPCA\n",
    "    pca = RandomizedPCA().fit(X_train)\n",
    "    plt.plot(np.cumsum(pca.explained_variance_ratio_))\n",
    "    plt.xlabel('number of components')\n",
    "    plt.ylabel('cumulative explained variance');\n",
    "    \n",
    "test_PCA(X_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "def apply_PCA(X_train,X_test):\n",
    "    from sklearn.decomposition import PCA as RandomizedPCA\n",
    "    pca = RandomizedPCA(n_components=30, whiten=True).fit(X_train)\n",
    "    X_train = pca.transform(X_train)\n",
    "    X_test = pca.transform(X_test)\n",
    "    \n",
    "    return X_train,X_test\n",
    "\n",
    "X_train_pca,X_test_pca = apply_PCA(X_train,X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(59475, 30)"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train_pca.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x20b5521a2b0>"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(15,10))\n",
    "sns.boxplot(data=X_train_pca)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "_________________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Machine learning models for prediction of risk factors"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "rf = RandomForestClassifier(n_estimators=750, random_state=45,bootstrap= False, criterion= 'entropy', n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=False, class_weight=None, criterion='entropy',\n",
       "                       max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
       "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                       min_samples_leaf=1, min_samples_split=2,\n",
       "                       min_weight_fraction_leaf=0.0, n_estimators=750,\n",
       "                       n_jobs=-1, oob_score=False, random_state=45, verbose=0,\n",
       "                       warm_start=False)"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf.fit(X_train_pca,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "rfpred = rf.predict(X_test_pca)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(14869,)"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfpred.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "labels = ['1','2','3','4','5','6','7','8']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           1       0.94      0.94      0.94      1871\n",
      "           2       0.89      0.95      0.92      1692\n",
      "           3       0.99      0.97      0.98      1896\n",
      "           4       0.98      0.91      0.94      1998\n",
      "           5       0.92      0.85      0.88      2011\n",
      "           6       0.73      0.80      0.76      1644\n",
      "           7       0.76      0.83      0.79      1797\n",
      "           8       0.81      0.78      0.79      1960\n",
      "\n",
      "    accuracy                           0.88     14869\n",
      "   macro avg       0.88      0.88      0.88     14869\n",
      "weighted avg       0.88      0.88      0.88     14869\n",
      "\n",
      "[[1753    8    3   12   29   19   17   28]\n",
      " [  11 1603   10   10   62   27   52   19]\n",
      " [   0    0 1838    6    2    3    2    0]\n",
      " [   0    2    4 1816    3    8    5   13]\n",
      " [  10   14    7   15 1700   39   23   44]\n",
      " [  38   24   22   73  113 1316   89  137]\n",
      " [  36   24    8   30   66  107 1487  189]\n",
      " [  23   17    4   36   36  125  122 1530]]\n"
     ]
    }
   ],
   "source": [
    "def create_model_report(pred,actual,labels):\n",
    "    from sklearn.metrics import classification_report, confusion_matrix\n",
    "    classification_report = classification_report(pred,actual)\n",
    "    conf_mat = confusion_matrix(y_true=actual, y_pred=pred)\n",
    "    fig = plt.figure()\n",
    "    ax = fig.add_subplot(111)\n",
    "    cax = ax.matshow(conf_mat, cmap=plt.cm.Blues)\n",
    "    fig.colorbar(cax)\n",
    "    ax.set_xticklabels([''] + labels)\n",
    "    ax.set_yticklabels([''] + labels)\n",
    "    plt.xlabel('Predicted')\n",
    "    plt.ylabel('Expected')\n",
    "    plt.show()\n",
    "    \n",
    "    print(classification_report) \n",
    "    print(conf_mat)\n",
    "\n",
    "create_model_report(rfpred,Y_test,labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Grid search the Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "grid_param = {\n",
    "    'n_estimators': [100, 300, 500, 750],\n",
    "    'criterion': ['gini', 'entropy'],\n",
    "    'bootstrap': [True, False]\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "gd_sr = GridSearchCV(estimator=rf,\n",
    "                     param_grid=grid_param,\n",
    "                     scoring='accuracy',\n",
    "                     cv=3,\n",
    "                     n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score='raise-deprecating',\n",
       "       estimator=RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
       "            max_depth=None, max_features='auto', max_leaf_nodes=None,\n",
       "            min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "            min_samples_leaf=1, min_samples_split=2,\n",
       "            min_weight_fraction_leaf=0.0, n_estimators=300, n_jobs=None,\n",
       "            oob_score=False, random_state=45, verbose=0, warm_start=False),\n",
       "       fit_params=None, iid='warn', n_jobs=-1,\n",
       "       param_grid={'n_estimators': [100, 300, 500], 'criterion': ['gini', 'entropy'], 'bootstrap': [True, False]},\n",
       "       pre_dispatch='2*n_jobs', refit=True, return_train_score='warn',\n",
       "       scoring='accuracy', verbose=0)"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gd_sr.fit(X_train_pca,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'bootstrap': False, 'criterion': 'entropy', 'n_estimators': 500}\n"
     ]
    }
   ],
   "source": [
    "best_parameters = gd_sr.best_params_\n",
    "print(best_parameters)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "_____________"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Support Vector Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.svm import SVC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "svmmodel = SVC(C=10, kernel = 'rbf',gamma=0.1, n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "SVC(C=10, cache_size=200, class_weight=None, coef0=0.0,\n",
       "    decision_function_shape='ovr', degree=3, gamma=0.1, kernel='rbf',\n",
       "    max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
       "    tol=0.001, verbose=False)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svmmodel.fit(X_train_pca,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "svmpred = svmmodel.predict(X_test_pca)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(14869,)"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svmpred.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           1       0.98      0.94      0.96      1935\n",
      "           2       0.96      0.94      0.95      1831\n",
      "           3       1.00      0.99      0.99      1877\n",
      "           4       1.00      0.96      0.98      1911\n",
      "           5       0.94      0.92      0.93      1882\n",
      "           6       0.80      0.84      0.82      1722\n",
      "           7       0.80      0.87      0.84      1797\n",
      "           8       0.82      0.81      0.82      1914\n",
      "\n",
      "    accuracy                           0.91     14869\n",
      "   macro avg       0.91      0.91      0.91     14869\n",
      "weighted avg       0.91      0.91      0.91     14869\n",
      "\n",
      "[[1823   10    1    5   10    8    7    5]\n",
      " [  14 1721    1    3   18   10   19    8]\n",
      " [   0    0 1851    0    0    0    0    0]\n",
      " [   2    1    1 1844    1    1    0    1]\n",
      " [   9   25    4    2 1736   31   24   21]\n",
      " [  34   30   11   37   61 1442   60  137]\n",
      " [  34   20    5   10   31   96 1567  184]\n",
      " [  19   24    3   10   25  134  120 1558]]\n"
     ]
    }
   ],
   "source": [
    "create_model_report(svmpred,Y_test,labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Grid search to find best parameters for SVM model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "Cs = [0.001, 0.01, 0.1, 1, 10]\n",
    "gammas = [0.001, 0.01, 0.1, 1,10]\n",
    "param_grid = {'C': Cs, 'gamma' : gammas}\n",
    "grid_search = GridSearchCV(svmmodel, param_grid = param_grid, cv=5,n_jobs=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score='raise-deprecating',\n",
       "       estimator=SVC(C=10, cache_size=200, class_weight=None, coef0=0.0,\n",
       "  decision_function_shape='ovr', degree=3, gamma=1, kernel='rbf',\n",
       "  max_iter=-1, probability=False, random_state=None, shrinking=True,\n",
       "  tol=0.001, verbose=False),\n",
       "       fit_params=None, iid='warn', n_jobs=None,\n",
       "       param_grid={'C': [0.001, 0.01, 0.1, 1, 10], 'gamma': [0.001, 0.01, 0.1, 1, 10]},\n",
       "       pre_dispatch='2*n_jobs', refit=True, return_train_score='warn',\n",
       "       scoring=None, verbose=0)"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_search.fit(X_train_pca,Y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'C': 10, 'gamma': 0.1}\n"
     ]
    }
   ],
   "source": [
    "best_params = grid_search.best_params_\n",
    "print(best_params)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "____________________"
   ]
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}