ds3.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "e2428246-9472-44e2-bffe-bee95f33028c",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import warnings\n",
    "import time\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from sklearn.model_selection import train_test_split, StratifiedKFold\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.discriminant_analysis import LinearDiscriminantAnalysis, QuadraticDiscriminantAnalysis\n",
    "from sklearn.metrics import recall_score, classification_report, mean_squared_error, mean_absolute_error, \\\n",
    "    accuracy_score, precision_score, f1_score, confusion_matrix, r2_score\n",
    "from xgboost import XGBRegressor\n",
    "from sklearn.datasets import make_regression\n",
    "from imblearn.over_sampling import SMOTE\n",
    "from imblearn.under_sampling import RandomUnderSampler\n",
    "from collections import Counter\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "6591c6b4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: xgboost in /opt/anaconda3/lib/python3.9/site-packages (2.0.3)\n",
      "Requirement already satisfied: scipy in /opt/anaconda3/lib/python3.9/site-packages (from xgboost) (1.13.0)\n",
      "Requirement already satisfied: numpy in /opt/anaconda3/lib/python3.9/site-packages (from xgboost) (1.26.4)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install xgboost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "7c2d624a-5f78-43d6-bfde-c73eef91dcd4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Numbers of Rows :  25134\n",
      "Number of columns :  20\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 25134 entries, 0 to 25133\n",
      "Data columns (total 20 columns):\n",
      " #   Column          Non-Null Count  Dtype  \n",
      "---  ------          --------------  -----  \n",
      " 0   Unnamed: 0      25134 non-null  int64  \n",
      " 1   ID              25134 non-null  int64  \n",
      " 2   GENDER          25134 non-null  object \n",
      " 3   CAR             25134 non-null  object \n",
      " 4   REALITY         25134 non-null  object \n",
      " 5   NO_OF_CHILD     25134 non-null  int64  \n",
      " 6   INCOME          25134 non-null  float64\n",
      " 7   INCOME_TYPE     25134 non-null  object \n",
      " 8   EDUCATION_TYPE  25134 non-null  object \n",
      " 9   FAMILY_TYPE     25134 non-null  object \n",
      " 10  HOUSE_TYPE      25134 non-null  object \n",
      " 11  FLAG_MOBIL      25134 non-null  int64  \n",
      " 12  WORK_PHONE      25134 non-null  int64  \n",
      " 13  PHONE           25134 non-null  int64  \n",
      " 14  E_MAIL          25134 non-null  int64  \n",
      " 15  FAMILY SIZE     25134 non-null  float64\n",
      " 16  BEGIN_MONTH     25134 non-null  int64  \n",
      " 17  AGE             25134 non-null  int64  \n",
      " 18  YEARS_EMPLOYED  25134 non-null  int64  \n",
      " 19  TARGET          25134 non-null  int64  \n",
      "dtypes: float64(2), int64(11), object(7)\n",
      "memory usage: 3.8+ MB\n",
      "None\n"
     ]
    },
    {
     "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>Unnamed: 0</th>\n",
       "      <th>ID</th>\n",
       "      <th>GENDER</th>\n",
       "      <th>CAR</th>\n",
       "      <th>REALITY</th>\n",
       "      <th>NO_OF_CHILD</th>\n",
       "      <th>INCOME</th>\n",
       "      <th>INCOME_TYPE</th>\n",
       "      <th>EDUCATION_TYPE</th>\n",
       "      <th>FAMILY_TYPE</th>\n",
       "      <th>HOUSE_TYPE</th>\n",
       "      <th>FLAG_MOBIL</th>\n",
       "      <th>WORK_PHONE</th>\n",
       "      <th>PHONE</th>\n",
       "      <th>E_MAIL</th>\n",
       "      <th>FAMILY SIZE</th>\n",
       "      <th>BEGIN_MONTH</th>\n",
       "      <th>AGE</th>\n",
       "      <th>YEARS_EMPLOYED</th>\n",
       "      <th>TARGET</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>5008806</td>\n",
       "      <td>M</td>\n",
       "      <td>Y</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>112500.0</td>\n",
       "      <td>Working</td>\n",
       "      <td>Secondary / secondary special</td>\n",
       "      <td>Married</td>\n",
       "      <td>House / apartment</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>29</td>\n",
       "      <td>59</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>5008808</td>\n",
       "      <td>F</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>270000.0</td>\n",
       "      <td>Commercial associate</td>\n",
       "      <td>Secondary / secondary special</td>\n",
       "      <td>Single / not married</td>\n",
       "      <td>House / apartment</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4</td>\n",
       "      <td>52</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>5008809</td>\n",
       "      <td>F</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>270000.0</td>\n",
       "      <td>Commercial associate</td>\n",
       "      <td>Secondary / secondary special</td>\n",
       "      <td>Single / not married</td>\n",
       "      <td>House / apartment</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>26</td>\n",
       "      <td>52</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>5008810</td>\n",
       "      <td>F</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>270000.0</td>\n",
       "      <td>Commercial associate</td>\n",
       "      <td>Secondary / secondary special</td>\n",
       "      <td>Single / not married</td>\n",
       "      <td>House / apartment</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>26</td>\n",
       "      <td>52</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>5008811</td>\n",
       "      <td>F</td>\n",
       "      <td>N</td>\n",
       "      <td>Y</td>\n",
       "      <td>0</td>\n",
       "      <td>270000.0</td>\n",
       "      <td>Commercial associate</td>\n",
       "      <td>Secondary / secondary special</td>\n",
       "      <td>Single / not married</td>\n",
       "      <td>House / apartment</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>38</td>\n",
       "      <td>52</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0       ID GENDER CAR REALITY  NO_OF_CHILD    INCOME  \\\n",
       "0           0  5008806      M   Y       Y            0  112500.0   \n",
       "1           1  5008808      F   N       Y            0  270000.0   \n",
       "2           2  5008809      F   N       Y            0  270000.0   \n",
       "3           3  5008810      F   N       Y            0  270000.0   \n",
       "4           4  5008811      F   N       Y            0  270000.0   \n",
       "\n",
       "            INCOME_TYPE                 EDUCATION_TYPE           FAMILY_TYPE  \\\n",
       "0               Working  Secondary / secondary special               Married   \n",
       "1  Commercial associate  Secondary / secondary special  Single / not married   \n",
       "2  Commercial associate  Secondary / secondary special  Single / not married   \n",
       "3  Commercial associate  Secondary / secondary special  Single / not married   \n",
       "4  Commercial associate  Secondary / secondary special  Single / not married   \n",
       "\n",
       "          HOUSE_TYPE  FLAG_MOBIL  WORK_PHONE  PHONE  E_MAIL  FAMILY SIZE  \\\n",
       "0  House / apartment           1           0      0       0          2.0   \n",
       "1  House / apartment           1           0      1       1          1.0   \n",
       "2  House / apartment           1           0      1       1          1.0   \n",
       "3  House / apartment           1           0      1       1          1.0   \n",
       "4  House / apartment           1           0      1       1          1.0   \n",
       "\n",
       "   BEGIN_MONTH  AGE  YEARS_EMPLOYED  TARGET  \n",
       "0           29   59               3       0  \n",
       "1            4   52               8       0  \n",
       "2           26   52               8       0  \n",
       "3           26   52               8       0  \n",
       "4           38   52               8       0  "
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df=pd.read_csv('credit_dataset.csv')\n",
    "missing=df.isnull().sum()\n",
    "\n",
    "print(\"Numbers of Rows : \", df.shape[0])\n",
    "print(\"Number of columns : \", df.shape[1])\n",
    "print(df.info())\n",
    "df.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "85afbfd4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Unnamed: 0        25134\n",
       "ID                25134\n",
       "GENDER                2\n",
       "CAR                   2\n",
       "REALITY               2\n",
       "NO_OF_CHILD           9\n",
       "INCOME              195\n",
       "INCOME_TYPE           5\n",
       "EDUCATION_TYPE        5\n",
       "FAMILY_TYPE           5\n",
       "HOUSE_TYPE            6\n",
       "FLAG_MOBIL            1\n",
       "WORK_PHONE            2\n",
       "PHONE                 2\n",
       "E_MAIL                2\n",
       "FAMILY SIZE          10\n",
       "BEGIN_MONTH          61\n",
       "AGE                  47\n",
       "YEARS_EMPLOYED       43\n",
       "TARGET                2\n",
       "dtype: int64"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "2a953658",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Dropping the the FLAG_MOBIL attribute as it has only 1 unique value\n",
    "\n",
    "df.drop(columns=['FLAG_MOBIL'], inplace=True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "f43e8369",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of Fraudulent Transaction :  422  = 1.6790005570144029\n",
      "Number of Valid Transaction :  24712 = 98.3209994429856\n"
     ]
    }
   ],
   "source": [
    "fraud_trans = df['TARGET'].value_counts()[1]\n",
    "valid_trans = df['TARGET'].value_counts()[0]\n",
    "\n",
    "print(\"Number of Fraudulent Transaction : \",fraud_trans,\" =\" ,(fraud_trans/25134)*100)\n",
    "print(\"Number of Valid Transaction : \" ,valid_trans,\"=\",(valid_trans/25134)*100)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "8574519b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "# Given values and labels\n",
    "values = [24712, 422]\n",
    "labels = ['0', '1']\n",
    "\n",
    "# Create a pie chart\n",
    "plt.figure(figsize=(6, 6))  # Optional: set the figure size\n",
    "plt.pie(values, labels=labels, autopct='%1.1f%%', startangle=140)\n",
    "\n",
    "# Add a title (optional)\n",
    "plt.title('Fraudulent vs Non-Fraudulent Transactions')\n",
    "\n",
    "# Display the pie chart\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "b04ba8eb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gender :  ['M' 'F']\n",
      "car :  ['Y' 'N']\n",
      "reality :  ['Y' 'N']\n",
      "income_type :  ['Working' 'Commercial associate' 'State servant' 'Student' 'Pensioner']\n",
      "education_type :  ['Secondary / secondary special' 'Higher education' 'Incomplete higher'\n",
      " 'Lower secondary' 'Academic degree']\n",
      "house_type :  ['House / apartment' 'Rented apartment' 'Municipal apartment'\n",
      " 'With parents' 'Co-op apartment' 'Office apartment']\n",
      "family_type :  ['Married' 'Single / not married' 'Civil marriage' 'Separated' 'Widow']\n"
     ]
    }
   ],
   "source": [
    "#Encoding the Attributes :\n",
    "'''\n",
    "gender”, “car”, “reality”,\n",
    "“income_type”, “education_type”, “house_type”, and “family_type '''\n",
    "\n",
    "print(\"Gender : \",df['GENDER'].unique())\n",
    "print(\"car : \",df['CAR'].unique())\n",
    "print(\"reality : \",df['REALITY'].unique())\n",
    "print(\"income_type : \",df['INCOME_TYPE'].unique())\n",
    "print(\"education_type : \",df['EDUCATION_TYPE'].unique())\n",
    "print(\"house_type : \",df['HOUSE_TYPE'].unique())\n",
    "print(\"family_type : \",df['FAMILY_TYPE'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 140,
   "id": "d3761370",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "\n",
    "columns_to_encode = [\"GENDER\", \"CAR\", \"REALITY\", \"FAMILY_TYPE\", \"INCOME_TYPE\", \"EDUCATION_TYPE\", \"HOUSE_TYPE\"]\n",
    "le = LabelEncoder()\n",
    "\n",
    "for column in columns_to_encode:\n",
    "    encoded = le.fit_transform(df[column])\n",
    "    df[column] = encoded\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "id": "75c7abe3",
   "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>Unnamed: 0</th>\n",
       "      <th>ID</th>\n",
       "      <th>GENDER</th>\n",
       "      <th>CAR</th>\n",
       "      <th>REALITY</th>\n",
       "      <th>NO_OF_CHILD</th>\n",
       "      <th>INCOME</th>\n",
       "      <th>INCOME_TYPE</th>\n",
       "      <th>EDUCATION_TYPE</th>\n",
       "      <th>FAMILY_TYPE</th>\n",
       "      <th>HOUSE_TYPE</th>\n",
       "      <th>WORK_PHONE</th>\n",
       "      <th>PHONE</th>\n",
       "      <th>E_MAIL</th>\n",
       "      <th>FAMILY SIZE</th>\n",
       "      <th>BEGIN_MONTH</th>\n",
       "      <th>AGE</th>\n",
       "      <th>YEARS_EMPLOYED</th>\n",
       "      <th>TARGET</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>508</th>\n",
       "      <td>508</td>\n",
       "      <td>5009628</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>238500</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>59</td>\n",
       "      <td>53</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>664</th>\n",
       "      <td>664</td>\n",
       "      <td>5009938</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>157500</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>39</td>\n",
       "      <td>29</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>956</th>\n",
       "      <td>956</td>\n",
       "      <td>5010535</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>135000</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>52</td>\n",
       "      <td>38</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>960</th>\n",
       "      <td>960</td>\n",
       "      <td>5010541</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>135000</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>52</td>\n",
       "      <td>38</td>\n",
       "      <td>8</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1177</th>\n",
       "      <td>1177</td>\n",
       "      <td>5010864</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>450000</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>51</td>\n",
       "      <td>50</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25129</th>\n",
       "      <td>25129</td>\n",
       "      <td>5149828</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>315000</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>48</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25130</th>\n",
       "      <td>25130</td>\n",
       "      <td>5149834</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>157500</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>23</td>\n",
       "      <td>34</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25131</th>\n",
       "      <td>25131</td>\n",
       "      <td>5149838</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>157500</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>32</td>\n",
       "      <td>34</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25132</th>\n",
       "      <td>25132</td>\n",
       "      <td>5150049</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>283500</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>9</td>\n",
       "      <td>49</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25133</th>\n",
       "      <td>25133</td>\n",
       "      <td>5150337</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>112500</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>25</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>422 rows × 19 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Unnamed: 0       ID  GENDER  CAR  REALITY  NO_OF_CHILD  INCOME  \\\n",
       "508           508  5009628       0    0        0            0  238500   \n",
       "664           664  5009938       0    0        1            2  157500   \n",
       "956           956  5010535       0    1        0            2  135000   \n",
       "960           960  5010541       0    1        0            2  135000   \n",
       "1177         1177  5010864       1    1        1            1  450000   \n",
       "...           ...      ...     ...  ...      ...          ...     ...   \n",
       "25129       25129  5149828       1    1        1            0  315000   \n",
       "25130       25130  5149834       0    0        1            0  157500   \n",
       "25131       25131  5149838       0    0        1            0  157500   \n",
       "25132       25132  5150049       0    0        1            0  283500   \n",
       "25133       25133  5150337       1    0        1            0  112500   \n",
       "\n",
       "       INCOME_TYPE  EDUCATION_TYPE  FAMILY_TYPE  HOUSE_TYPE  WORK_PHONE  \\\n",
       "508              4               4            1           1           1   \n",
       "664              4               4            1           1           0   \n",
       "956              4               4            1           1           0   \n",
       "960              4               4            1           1           0   \n",
       "1177             0               4            1           1           0   \n",
       "...            ...             ...          ...         ...         ...   \n",
       "25129            4               4            1           1           0   \n",
       "25130            0               1            1           1           0   \n",
       "25131            1               1            1           1           0   \n",
       "25132            4               4            1           1           0   \n",
       "25133            4               4            3           4           0   \n",
       "\n",
       "       PHONE  E_MAIL  FAMILY SIZE  BEGIN_MONTH  AGE  YEARS_EMPLOYED  TARGET  \n",
       "508        0       0            2           59   53               9       1  \n",
       "664        0       0            4           39   29               6       1  \n",
       "956        0       0            4           52   38               8       1  \n",
       "960        0       0            4           52   38               8       1  \n",
       "1177       1       1            3           51   50               2       1  \n",
       "...      ...     ...          ...          ...  ...             ...     ...  \n",
       "25129      0       0            2           11   48               7       1  \n",
       "25130      1       1            2           23   34               4       1  \n",
       "25131      1       1            2           32   34               4       1  \n",
       "25132      0       0            2            9   49               2       1  \n",
       "25133      0       0            1           13   25               3       1  \n",
       "\n",
       "[422 rows x 19 columns]"
      ]
     },
     "execution_count": 141,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rows_with_target_1 = df[df['TARGET'] == 1]\n",
    "rows_with_target_1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "f85a4a29",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['INCOME'] = df['INCOME'].astype('int64')\n",
    "df['FAMILY SIZE'] = df['FAMILY SIZE'].astype('int64')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "9764b2e9-d1e0-4153-9751-9c4123ca0089",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 25134 entries, 0 to 25133\n",
      "Data columns (total 18 columns):\n",
      " #   Column          Non-Null Count  Dtype\n",
      "---  ------          --------------  -----\n",
      " 0   Unnamed: 0      25134 non-null  int64\n",
      " 1   ID              25134 non-null  int64\n",
      " 2   GENDER          25134 non-null  int64\n",
      " 3   CAR             25134 non-null  int64\n",
      " 4   REALITY         25134 non-null  int64\n",
      " 5   NO_OF_CHILD     25134 non-null  int64\n",
      " 6   INCOME          25134 non-null  int64\n",
      " 7   INCOME_TYPE     25134 non-null  int64\n",
      " 8   EDUCATION_TYPE  25134 non-null  int64\n",
      " 9   FAMILY_TYPE     25134 non-null  int64\n",
      " 10  HOUSE_TYPE      25134 non-null  int64\n",
      " 11  WORK_PHONE      25134 non-null  int64\n",
      " 12  PHONE           25134 non-null  int64\n",
      " 13  E_MAIL          25134 non-null  int64\n",
      " 14  FAMILY SIZE     25134 non-null  int64\n",
      " 15  BEGIN_MONTH     25134 non-null  int64\n",
      " 16  AGE             25134 non-null  int64\n",
      " 17  YEARS_EMPLOYED  25134 non-null  int64\n",
      "dtypes: int64(18)\n",
      "memory usage: 3.5 MB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "X=df.drop(columns='TARGET')\n",
    "y=df['TARGET']\n",
    "\n",
    "print(X.info())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "5082b24f",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "X_train, X_test, y_train, y_test= train_test_split(X, y, test_size= 0.25, random_state=0)  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "265122aa",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Time taken to train the model: 0.23 seconds\n"
     ]
    }
   ],
   "source": [
    "# Using dt classifier \n",
    "start_time = time.time()\n",
    "clf = DecisionTreeClassifier()\n",
    "clf = clf.fit(X_train,y_train)\n",
    "y_pred = clf.predict(X_test)\n",
    "\n",
    "end_time = time.time()\n",
    "training_time = end_time - start_time\n",
    "print(f\"Time taken to train the model: {training_time:.2f} seconds\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "5064ddbd",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6728971962616822"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recall = recall_score(y_test, y_pred)\n",
    "recall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "f07cc7db",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.99      0.99      0.99      6177\n",
      "           1       0.60      0.67      0.63       107\n",
      "\n",
      "    accuracy                           0.99      6284\n",
      "   macro avg       0.79      0.83      0.81      6284\n",
      "weighted avg       0.99      0.99      0.99      6284\n",
      "\n"
     ]
    }
   ],
   "source": [
    "report = classification_report(y_test, y_pred)\n",
    "print(report)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "6727b3d8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.99      1.00      1.00      6177\n",
      "           1       0.87      0.67      0.76       107\n",
      "\n",
      "    accuracy                           0.99      6284\n",
      "   macro avg       0.93      0.84      0.88      6284\n",
      "weighted avg       0.99      0.99      0.99      6284\n",
      "\n",
      "Time taken to train the model: 3.12 seconds\n"
     ]
    }
   ],
   "source": [
    "start_time = time.time()\n",
    "rf_clf = RandomForestClassifier(random_state=0)\n",
    "rf_clf.fit(X_train, y_train)\n",
    "y_pred = rf_clf.predict(X_test)\n",
    "print(classification_report(y_test, y_pred))\n",
    "end_time = time.time()\n",
    "training_time = end_time - start_time\n",
    "\n",
    "\n",
    "print(f\"Time taken to train the model: {training_time:.2f} seconds\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "2727b1ba",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Classification Report:\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.99      0.99      0.99      6177\n",
      "           1       0.32      0.39      0.35       107\n",
      "\n",
      "    accuracy                           0.98      6284\n",
      "   macro avg       0.66      0.69      0.67      6284\n",
      "weighted avg       0.98      0.98      0.98      6284\n",
      "\n"
     ]
    }
   ],
   "source": [
    "qda_model = QuadraticDiscriminantAnalysis()\n",
    "qda_model.fit(X_train, y_train)\n",
    "y_pred = qda_model.predict(X_test)\n",
    "print(\"Classification Report:\")\n",
    "print(classification_report(y_test, y_pred))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "d38d697d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: imblearn in /opt/anaconda3/lib/python3.9/site-packages (0.0)\n",
      "Requirement already satisfied: imbalanced-learn in /opt/anaconda3/lib/python3.9/site-packages (from imblearn) (0.12.0)\n",
      "Requirement already satisfied: threadpoolctl>=2.0.0 in /opt/anaconda3/lib/python3.9/site-packages (from imbalanced-learn->imblearn) (2.2.0)\n",
      "Requirement already satisfied: scipy>=1.5.0 in /opt/anaconda3/lib/python3.9/site-packages (from imbalanced-learn->imblearn) (1.13.0)\n",
      "Requirement already satisfied: numpy>=1.17.3 in /opt/anaconda3/lib/python3.9/site-packages (from imbalanced-learn->imblearn) (1.26.4)\n",
      "Requirement already satisfied: joblib>=1.1.1 in /opt/anaconda3/lib/python3.9/site-packages (from imbalanced-learn->imblearn) (1.3.2)\n",
      "Requirement already satisfied: scikit-learn>=1.0.2 in /opt/anaconda3/lib/python3.9/site-packages (from imbalanced-learn->imblearn) (1.4.1.post1)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install imblearn\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "aecc0bbd-dce9-468c-ac9a-48a968130281",
   "metadata": {},
   "outputs": [],
   "source": [
    "smote=SMOTE(sampling_strategy='minority')\n",
    "x_sm,y_sm=smote.fit_resample(X,y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "0846245a-58a7-4ce8-ab91-8714873f962e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TARGET\n",
       "0    24712\n",
       "1    24712\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_sm.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "ec81e8a1-6da6-4692-825c-e2e88052e126",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TARGET\n",
       "0    24712\n",
       "1    24712\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_sm.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "dbfa7449",
   "metadata": {},
   "outputs": [],
   "source": [
    "def underSampling():\n",
    "    \n",
    "\n",
    "    undersampler = RandomUnderSampler(sampling_strategy='auto', random_state=42)\n",
    "    x_under,y_under = undersampler.fit_resample(X,y)\n",
    "    \n",
    "   \n",
    "    xtrain,xtest,ytrain,ytest=train_test_split(x_under,y_under,test_size=.20,stratify=y_under)\n",
    "    \n",
    "    knn = KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski', metric_params=None, n_jobs=-1, n_neighbors=5, p=2, weights='uniform')\n",
    "\n",
    "    knn.fit(xtrain,ytrain)\n",
    "    ypre_knn=knn.predict(xtest)\n",
    "    print(\"report\\n\",classification_report(ytest,ypre_knn,digits=6))\n",
    "    \n",
    "    lda = LinearDiscriminantAnalysis(covariance_estimator=None, n_components=None, priors=None, shrinkage=None, solver='svd', store_covariance=False, tol=0.0001)\n",
    "\n",
    "    lda.fit(xtrain, ytrain)\n",
    "    ypre_lda = lda.predict(xtest)\n",
    "    print(\"lda report\\n\", classification_report(ytest, ypre_lda))\n",
    "    \n",
    "    lr=LinearRegression()\n",
    "    lr.fit(xtrain,ytrain)\n",
    "    ypre_lr=lr.predict(xtest)\n",
    "\n",
    "    mse = mean_squared_error(ytest, ypre_lr)\n",
    "    mae = mean_absolute_error(ytest, ypre_lr)\n",
    "\n",
    "\n",
    "    print(\"Mean Squared Error (MSE):\", mse)\n",
    "    print(\"Mean Absolute Error (MAE):\", mae)\n",
    "    \n",
    "    mvl = np.mean(ypre_lr)\n",
    "\n",
    "    # Initialize the final predictions array\n",
    "    por = np.zeros_like(ytest)\n",
    "\n",
    "    for i in range(0,len(ypre_lr)):\n",
    "        if(ypre_knn[i]==0 or ypre_lda[i]==0):\n",
    "            if(ypre_lr[i]<mvl):\n",
    "                por[i]=0\n",
    "        elif(ypre_knn[i]==1 or ypre_lda[i]==1):\n",
    "            if(ypre_lr[i]>mvl):\n",
    "                por[i]=1\n",
    "        else:\n",
    "            por[i]=ypre_knn[i]\n",
    "    print(classification_report(ytest,por))\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "af761807",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "report\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0   0.758242  0.811765  0.784091        85\n",
      "           1   0.794872  0.738095  0.765432        84\n",
      "\n",
      "    accuracy                       0.775148       169\n",
      "   macro avg   0.776557  0.774930  0.774762       169\n",
      "weighted avg   0.776448  0.775148  0.774817       169\n",
      "\n",
      "lda report\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       0.82      0.80      0.81        85\n",
      "           1       0.80      0.82      0.81        84\n",
      "\n",
      "    accuracy                           0.81       169\n",
      "   macro avg       0.81      0.81      0.81       169\n",
      "weighted avg       0.81      0.81      0.81       169\n",
      "\n",
      "Mean Squared Error (MSE): 0.1316675044582883\n",
      "Mean Absolute Error (MAE): 0.307717330025218\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.71      0.93      0.81        85\n",
      "           1       0.90      0.62      0.73        84\n",
      "\n",
      "    accuracy                           0.78       169\n",
      "   macro avg       0.80      0.77      0.77       169\n",
      "weighted avg       0.80      0.78      0.77       169\n",
      "\n"
     ]
    }
   ],
   "source": [
    "underSampling()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "1d4f5091",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "categories = ['OverSampling', \"UnderSampling\"]\n",
    "values = [0.95,0.64 ]  \n",
    "colors = ['blue', 'blue']\n",
    "fig, ax = plt.subplots()\n",
    "bars = ax.bar(categories, values, color=colors)\n",
    "for bar in bars:\n",
    "    x_val = bar.get_x() + bar.get_width() / 2  \n",
    "    y_val = bar.get_height()  \n",
    "    ax.text(x_val, y_val, f'{y_val:.2f}',  \n",
    "            ha='center', va='bottom')  \n",
    "\n",
    "ax.set_title('Recall for dataset 3')\n",
    "ax.set_xlabel('Handling imbalanced dataset ')\n",
    "ax.set_ylabel('Recall values')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "dd4163e7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Target , Age , Family_Size , Phone , No_of_Child , Unnamed \n",
    "\n",
    "#Correlation matrix\n",
    "\n",
    "\n",
    "\n",
    "# Select the columns you want to include in the correlation matrix\n",
    "columns = ['TARGET', 'AGE', 'FAMILY SIZE', 'PHONE', 'NO_OF_CHILD']\n",
    "\n",
    "# Compute the correlation matrix\n",
    "corr_matrix = df[columns].corr()\n",
    "\n",
    "# Plot the heatmap\n",
    "plt.figure(figsize=(10, 8))\n",
    "sns.heatmap(corr_matrix, annot=True, cmap='coolwarm', fmt=\".2f\")  # fmt=\".2f\" for 2 decimal places\n",
    "plt.title('Correlation Matrix')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "0748cc15",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Unnamed: 0', 'ID', 'GENDER', 'CAR', 'REALITY', 'NO_OF_CHILD', 'INCOME', 'INCOME_TYPE', 'EDUCATION_TYPE', 'FAMILY_TYPE', 'HOUSE_TYPE', 'WORK_PHONE', 'PHONE', 'E_MAIL', 'FAMILY SIZE', 'BEGIN_MONTH', 'AGE', 'YEARS_EMPLOYED']\n",
      "18\n"
     ]
    }
   ],
   "source": [
    "column_names = list(df.columns)\n",
    "\n",
    "column_names.remove(\"TARGET\")\n",
    "\n",
    "print(column_names)\n",
    "print(len(column_names))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "86993cfa",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Unnamed: 0: 0.13839\n",
      "ID: 0.00104\n",
      "GENDER: 0.00154\n",
      "CAR: 0.00199\n",
      "REALITY: 0.00109\n",
      "NO_OF_CHILD: 0.30016\n",
      "INCOME: 0.00058\n",
      "INCOME_TYPE: 0.00115\n",
      "EDUCATION_TYPE: 0.53974\n",
      "FAMILY_TYPE: 0.00191\n",
      "HOUSE_TYPE: 0.00220\n",
      "FLAG_MOBIL: 0.00102\n",
      "WORK_PHONE: 0.00224\n",
      "PHONE: 0.00072\n",
      "E_MAIL: 0.00280\n",
      "FAMILY SIZE: 0.00092\n",
      "BEGIN_MONTH: 0.00080\n",
      "AGE: 0.00081\n",
      "YEARS_EMPLOYED: 0.00089\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "\n",
    "\n",
    "# Define dataset\n",
    "X, y = make_regression(n_samples=1000, n_features=19, n_informative=5, random_state=1)\n",
    "\n",
    "# Define the model\n",
    "model = XGBRegressor()\n",
    "\n",
    "# Fit the model\n",
    "model.fit(X, y)\n",
    "\n",
    "# Get importance\n",
    "importance = model.feature_importances_\n",
    "\n",
    "# Get feature names from the dataset (assuming you have them)\n",
    "feature_names = ['Unnamed: 0', 'ID', 'GENDER', 'CAR', 'REALITY', 'NO_OF_CHILD', 'INCOME', 'INCOME_TYPE', 'EDUCATION_TYPE', 'FAMILY_TYPE', 'HOUSE_TYPE', 'FLAG_MOBIL', 'WORK_PHONE', 'PHONE', 'E_MAIL', 'FAMILY SIZE', 'BEGIN_MONTH', 'AGE', 'YEARS_EMPLOYED'] # Modify this based on your actual feature names\n",
    "\n",
    "# Summarize feature importance\n",
    "for i,v in enumerate(importance):\n",
    "    print(f'{feature_names[i]}: {v:.5f}')\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(10, 6))  # Adjust figure size as needed\n",
    "bar = ax.bar(feature_names, importance)\n",
    "ax.set_xlabel('Features')\n",
    "ax.set_ylabel('Importance Score')\n",
    "ax.set_title('Feature Importance')\n",
    "ax.tick_params(axis='x', rotation=45)  # Rotate x-axis labels for better readability\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "22ac657c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EDUCATION_TYPE: 0.53974\n",
      "NO_OF_CHILD: 0.30016\n",
      "Unnamed: 0: 0.13839\n",
      "E_MAIL: 0.0028\n",
      "WORK_PHONE: 0.00224\n",
      "HOUSE_TYPE: 0.0022\n",
      "CAR: 0.00199\n",
      "FAMILY_TYPE: 0.00191\n",
      "GENDER: 0.00154\n",
      "INCOME_TYPE: 0.00115\n",
      "REALITY: 0.00109\n",
      "ID: 0.00104\n",
      "FLAG_MOBIL: 0.00102\n",
      "FAMILY SIZE: 0.00092\n",
      "YEARS_EMPLOYED: 0.00089\n",
      "AGE: 0.00081\n",
      "BEGIN_MONTH: 0.0008\n",
      "PHONE: 0.00072\n",
      "INCOME: 0.00058\n"
     ]
    }
   ],
   "source": [
    "# Original attribute-value pairs\n",
    "attributes = {\n",
    "    'Unnamed: 0': 0.13839,\n",
    "    'ID': 0.00104,\n",
    "    'GENDER': 0.00154,\n",
    "    'CAR': 0.00199,\n",
    "    'REALITY': 0.00109,\n",
    "    'NO_OF_CHILD': 0.30016,\n",
    "    'INCOME': 0.00058,\n",
    "    'INCOME_TYPE': 0.00115,\n",
    "    'EDUCATION_TYPE': 0.53974,\n",
    "    'FAMILY_TYPE': 0.00191,\n",
    "    'HOUSE_TYPE': 0.00220,\n",
    "    'FLAG_MOBIL': 0.00102,\n",
    "    'WORK_PHONE': 0.00224,\n",
    "    'PHONE': 0.00072,\n",
    "    'E_MAIL': 0.00280,\n",
    "    'FAMILY SIZE': 0.00092,\n",
    "    'BEGIN_MONTH': 0.00080,\n",
    "    'AGE': 0.00081,\n",
    "    'YEARS_EMPLOYED': 0.00089\n",
    "}\n",
    "\n",
    "# Sort the attributes in descending order of values\n",
    "sorted_attributes = sorted(attributes.items(), key=lambda x: x[1], reverse=True)\n",
    "\n",
    "# Print the sorted attributes\n",
    "for attr, value in sorted_attributes:\n",
    "    print(f\"{attr}: {value}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "9189f8f7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top 5 Attributes:\n",
      "Mean Recall (kNN): 0.721390439435301\n",
      "Mean Recall (LDA): 0.7875526939935991\n",
      "\n",
      "Top 6 Attributes:\n",
      "Mean Recall (kNN): 0.7262059186584612\n",
      "Mean Recall (LDA): 0.7886452052794819\n",
      "\n",
      "Top 7 Attributes:\n",
      "Mean Recall (kNN): 0.7291195140588136\n",
      "Mean Recall (LDA): 0.8098898302649393\n",
      "\n",
      "Top 8 Attributes:\n",
      "Mean Recall (kNN): 0.7352299664168281\n",
      "Mean Recall (LDA): 0.8034153821390644\n",
      "\n",
      "Top 9 Attributes:\n",
      "Mean Recall (kNN): 0.7372937034602399\n",
      "Mean Recall (LDA): 0.809242425569747\n",
      "\n",
      "Top 10 Attributes:\n",
      "Mean Recall (kNN): 0.7504453644882293\n",
      "Mean Recall (LDA): 0.7994094801334157\n",
      "\n",
      "Top 11 Attributes:\n",
      "Mean Recall (kNN): 0.7521044234504022\n",
      "Mean Recall (LDA): 0.8034963619663188\n",
      "\n",
      "Top 12 Attributes:\n",
      "Mean Recall (kNN): 0.9435493889752322\n",
      "Mean Recall (LDA): 0.7522663831049112\n",
      "\n",
      "Top 13 Attributes:\n",
      "Mean Recall (kNN): 0.9436707809374912\n",
      "Mean Recall (LDA): 0.7567984861496331\n",
      "\n",
      "Top 14 Attributes:\n",
      "Mean Recall (kNN): 0.9456940649097814\n",
      "Mean Recall (LDA): 0.7580530144005893\n",
      "\n",
      "Top 15 Attributes:\n",
      "Mean Recall (kNN): 0.9481625619066667\n",
      "Mean Recall (LDA): 0.7587409376646911\n",
      "\n",
      "Top 16 Attributes:\n",
      "Mean Recall (kNN): 0.9533826619005017\n",
      "Mean Recall (LDA): 0.8272097131909188\n",
      "\n",
      "Top 17 Attributes:\n",
      "Mean Recall (kNN): 0.9533826619005017\n",
      "Mean Recall (LDA): 0.8283426693606998\n",
      "\n",
      "Top 18 Attributes:\n",
      "Mean Recall (kNN): 0.9528161797219997\n",
      "Mean Recall (LDA): 0.8273310805915072\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Define the top N significant attributes\n",
    "significant_attributes = {\n",
    "    'Unnamed: 0': 0.13839,\n",
    "    'ID': 0.00104,\n",
    "    'GENDER': 0.00154,\n",
    "    'CAR': 0.00199,\n",
    "    'REALITY': 0.00109,\n",
    "    'NO_OF_CHILD': 0.30016,\n",
    "    'INCOME': 0.00058,\n",
    "    'INCOME_TYPE': 0.00115,\n",
    "    'EDUCATION_TYPE': 0.53974,\n",
    "    'FAMILY_TYPE': 0.00191,\n",
    "    'HOUSE_TYPE': 0.00220,\n",
    "    'WORK_PHONE': 0.00224,\n",
    "    'PHONE': 0.00072,\n",
    "    'E_MAIL': 0.00280,\n",
    "    'FAMILY SIZE': 0.00092,\n",
    "    'BEGIN_MONTH': 0.00080,\n",
    "    'AGE': 0.00081,\n",
    "    'YEARS_EMPLOYED': 0.00089\n",
    "}\n",
    "\n",
    "# Sort the attributes based on their importance\n",
    "sorted_attributes = sorted(significant_attributes.items(), key=lambda x: x[1], reverse=True)\n",
    "\n",
    "# Define lists to store recall values for kNN, LDA, and LR\n",
    "recall_knn_list = []\n",
    "recall_lda_list = []\n",
    "mse_LR = [] \n",
    "mae_LR = [] \n",
    "rmse_LR = []\n",
    "\n",
    "pKNN,pLDA,pLR =[],[],[]\n",
    "\n",
    "# Create StratifiedKFold object\n",
    "skf = StratifiedKFold(n_splits=5, shuffle=True, random_state=1)\n",
    "\n",
    "# Iterate over the range of top_n_attributes from 5 to 18\n",
    "for top_n_attributes in range(5, 19):\n",
    "    # Select the top N attributes\n",
    "    top_n_attributes_names = [attr for attr, _ in sorted_attributes[:top_n_attributes]]\n",
    "\n",
    "    # Initialize lists to store recall values for each fold\n",
    "    recall_knn_fold = []\n",
    "    recall_lda_fold = []\n",
    "    recall_lr_fold = []\n",
    "\n",
    "    # Iterate over the folds\n",
    "    for train_index, test_index in skf.split(x_sm, y_sm):\n",
    "        x_train_fold, x_test_fold = x_sm[top_n_attributes_names].iloc[train_index], x_sm[top_n_attributes_names].iloc[test_index]\n",
    "        y_train_fold, y_test_fold = y_sm.iloc[train_index], y_sm.iloc[test_index]\n",
    "\n",
    "        # Fit kNN model\n",
    "        knn = KNeighborsClassifier()\n",
    "        knn.fit(x_train_fold, y_train_fold)\n",
    "        y_pred_knn = knn.predict(x_test_fold)\n",
    "        pKNN.extend(y_pred_knn)\n",
    "        recall_knn = recall_score(y_test_fold, y_pred_knn)\n",
    "        recall_knn_fold.append(recall_knn)\n",
    "\n",
    "        # Fit LDA model\n",
    "        lda = LinearDiscriminantAnalysis()\n",
    "        lda.fit(x_train_fold, y_train_fold)\n",
    "        y_pred_lda = lda.predict(x_test_fold)\n",
    "        pLDA.extend(y_pred_lda)\n",
    "        recall_lda = recall_score(y_test_fold, y_pred_lda)\n",
    "        recall_lda_fold.append(recall_lda)\n",
    "\n",
    "        # Fit LR model\n",
    "        LR = LinearRegression()\n",
    "        LR.fit(x_train_fold, y_train_fold)\n",
    "        y_LR = LR.predict(x_test_fold)\n",
    "        pLR.extend(y_LR)\n",
    "        mse = mean_squared_error(y_test_fold, y_LR)\n",
    "        mae = mean_absolute_error(y_test_fold, y_LR)\n",
    "        rmse = np.sqrt(mse)\n",
    "\n",
    "        mse_LR.append(mse)\n",
    "        mae_LR.append(mae)\n",
    "        rmse_LR.append(rmse)\n",
    "\n",
    "    # Calculate the mean recall values for each model and append to the respective lists\n",
    "    recall_knn_list.append(np.mean(recall_knn_fold))\n",
    "    recall_lda_list.append(np.mean(recall_lda_fold))\n",
    "\n",
    "# Print the recall values for each top_n_attributes\n",
    "for i, top_n_attributes in enumerate(range(5, 19)):\n",
    "    print(f\"Top {top_n_attributes} Attributes:\")\n",
    "    print(f\"Mean Recall (kNN): {recall_knn_list[i]}\")\n",
    "    print(f\"Mean Recall (LDA): {recall_lda_list[i]}\")\n",
    "#     print(f\"Mean Recall (LR): {recall_lr_list[i]}\")\n",
    "    print()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "31ca4e9a",
   "metadata": {},
   "outputs": [],
   "source": [
    "xtrain,xtest,ytrain,ytest=train_test_split(x_sm,y_sm,test_size=.20,stratify=y_sm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "id": "257e4131",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Index: 9885 entries, 30591 to 33265\n",
      "Data columns (total 18 columns):\n",
      " #   Column          Non-Null Count  Dtype\n",
      "---  ------          --------------  -----\n",
      " 0   Unnamed: 0      9885 non-null   int64\n",
      " 1   ID              9885 non-null   int64\n",
      " 2   GENDER          9885 non-null   int64\n",
      " 3   CAR             9885 non-null   int64\n",
      " 4   REALITY         9885 non-null   int64\n",
      " 5   NO_OF_CHILD     9885 non-null   int64\n",
      " 6   INCOME          9885 non-null   int64\n",
      " 7   INCOME_TYPE     9885 non-null   int64\n",
      " 8   EDUCATION_TYPE  9885 non-null   int64\n",
      " 9   FAMILY_TYPE     9885 non-null   int64\n",
      " 10  HOUSE_TYPE      9885 non-null   int64\n",
      " 11  WORK_PHONE      9885 non-null   int64\n",
      " 12  PHONE           9885 non-null   int64\n",
      " 13  E_MAIL          9885 non-null   int64\n",
      " 14  FAMILY SIZE     9885 non-null   int64\n",
      " 15  BEGIN_MONTH     9885 non-null   int64\n",
      " 16  AGE             9885 non-null   int64\n",
      " 17  YEARS_EMPLOYED  9885 non-null   int64\n",
      "dtypes: int64(18)\n",
      "memory usage: 1.4 MB\n"
     ]
    }
   ],
   "source": [
    "xtest.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "8ce4c0b3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "report\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0   0.955709  0.982396  0.968868      4942\n",
      "           1   0.981894  0.954481  0.967993      4943\n",
      "\n",
      "    accuracy                       0.968437      9885\n",
      "   macro avg   0.968801  0.968438  0.968431      9885\n",
      "weighted avg   0.968803  0.968437  0.968431      9885\n",
      "\n",
      "Time taken to train the model: 17.12 seconds\n"
     ]
    }
   ],
   "source": [
    "#kNN\n",
    "\n",
    "start_time = time.time()\n",
    "knn = KNeighborsClassifier(algorithm='auto', leaf_size=30, metric='minkowski', metric_params=None, n_jobs=-1, n_neighbors=5, p=2, weights='uniform')\n",
    "knn.fit(xtrain,ytrain)\n",
    "ypre_knn=knn.predict(xtest)\n",
    "print(\"report\\n\",classification_report(ytest,ypre_knn,digits=6))\n",
    "end_time = time.time()\n",
    "training_time = end_time - start_time\n",
    "print(f\"Time taken to train the model: {training_time:.2f} seconds\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "81350344",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "counts = [len(ypre_knn[ypre_knn == 0]), len(ypre_knn[ypre_knn == 1])]\n",
    "labels = ['0 Prediction', '1 Prediction']\n",
    "plt.figure(figsize=(8, 8))\n",
    "plt.pie(counts, labels=labels, autopct='%1.1f%%', startangle=140)\n",
    "plt.title('Distribution of Predictions')\n",
    "plt.axis('equal') \n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "25b6e0d7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "lda report\n",
      "               precision    recall  f1-score   support\n",
      "\n",
      "           0       0.84      0.86      0.85      4942\n",
      "           1       0.86      0.83      0.84      4943\n",
      "\n",
      "    accuracy                           0.85      9885\n",
      "   macro avg       0.85      0.85      0.85      9885\n",
      "weighted avg       0.85      0.85      0.85      9885\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#LDA \n",
    "lda = LinearDiscriminantAnalysis(covariance_estimator=None, n_components=None, priors=None, shrinkage=None, solver='svd', store_covariance=False, tol=0.0001)\n",
    "lda.fit(xtrain, ytrain)\n",
    "ypre_lda = lda.predict(xtest)\n",
    "print(\"lda report\\n\", classification_report(ytest, ypre_lda))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "34895217",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean Squared Error (MSE): 0.12060286498371889\n",
      "Mean Absolute Error (MAE): 0.289129437857957\n"
     ]
    }
   ],
   "source": [
    "#LR \n",
    "lr=LinearRegression()\n",
    "lr.fit(xtrain,ytrain)\n",
    "ypre_lr=lr.predict(xtest)\n",
    "\n",
    "mse = mean_squared_error(ytest, ypre_lr)\n",
    "mae = mean_absolute_error(ytest, ypre_lr)\n",
    "\n",
    "\n",
    "print(\"Mean Squared Error (MSE):\", mse)\n",
    "print(\"Mean Absolute Error (MAE):\", mae)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "09a79734",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate the mean prediction of LR model\n",
    "\n",
    "mvl = np.mean(ypre_lr)\n",
    "\n",
    "# Initialize the final predictions array\n",
    "por = np.zeros_like(ytest)\n",
    "for i in range(0,len(ypre_lr)):\n",
    "    if(ypre_knn[i]==0 or ypre_lda[i]==0):\n",
    "        if(ypre_lr[i]<mvl):\n",
    "            por[i]=0\n",
    "    elif(ypre_knn[i]==1 or ypre_lda[i]==1):\n",
    "        if(ypre_lr[i]>mvl):\n",
    "            por[i]=1\n",
    "    else:\n",
    "        por[i]=ypre_knn[i]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "5c6197be",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.83      1.00      0.91      4942\n",
      "           1       1.00      0.80      0.89      4943\n",
      "\n",
      "    accuracy                           0.90      9885\n",
      "   macro avg       0.91      0.90      0.90      9885\n",
      "weighted avg       0.91      0.90      0.90      9885\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(ytest,por))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "3eebb544",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.84      0.87      0.85      4942\n",
      "           1       0.86      0.83      0.85      4943\n",
      "\n",
      "    accuracy                           0.85      9885\n",
      "   macro avg       0.85      0.85      0.85      9885\n",
      "weighted avg       0.85      0.85      0.85      9885\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Calculate the mean prediction of LR model\n",
    "mvl = np.mean(ypre_lr)\n",
    "por = np.zeros_like(ytest)\n",
    "for i in range(len(ypre_lr)):\n",
    "    if (ypre_knn[i] == 0 or ypre_lda[i] == 0) and ypre_lr[i] < mvl:\n",
    "        por[i] = 0\n",
    "    elif (ypre_knn[i] == 1 or ypre_lda[i] == 1) and ypre_lr[i] > mvl:\n",
    "        por[i] = 1\n",
    "    else:\n",
    "        por[i] = ypre_knn[i]\n",
    "\n",
    "print(classification_report(ytest,por))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "11c7d92c",
   "metadata": {},
   "outputs": [],
   "source": [
    "#modified ALgorithm - Novelity\n",
    "\n",
    "\n",
    "PKNN = ypre_knn\n",
    "PLDA = ypre_lda\n",
    "PLR = ypre_lr\n",
    "por = np.zeros_like(ytest)\n",
    "threshold = 0.43\n",
    "\n",
    "\n",
    "for i in range(len(ytest)):\n",
    "    avg = (PKNN[i] + PLDA[i] + PLR[i]) / 3\n",
    "    if avg > threshold:\n",
    "        por[i] = 1\n",
    "    else:\n",
    "        por[i] = 0\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "94fe5474",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      0.86      0.90      4942\n",
      "           1       0.87      0.96      0.91      4943\n",
      "\n",
      "    accuracy                           0.91      9885\n",
      "   macro avg       0.91      0.91      0.91      9885\n",
      "weighted avg       0.91      0.91      0.91      9885\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(ytest,por))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "609648dd",
   "metadata": {},
   "outputs": [],
   "source": [
    "#modified ALgorithm - Novelity\n",
    "# Here we have taken the threshold value as 0.5 \n",
    "\n",
    "PKNN = ypre_knn\n",
    "PLDA = ypre_lda\n",
    "PLR = ypre_lr\n",
    "por = np.zeros_like(ytest)\n",
    "threshold = 0.5\n",
    "for i in range(len(ytest)):\n",
    "\n",
    "    avg = (PKNN[i] + PLDA[i] + PLR[i]) / 3\n",
    "    if avg > threshold:\n",
    "        por[i] = 1\n",
    "    else:\n",
    "        por[i] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "7c6b621e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.84      0.86      0.85      4942\n",
      "           1       0.86      0.83      0.84      4943\n",
      "\n",
      "    accuracy                           0.85      9885\n",
      "   macro avg       0.85      0.85      0.85      9885\n",
      "weighted avg       0.85      0.85      0.85      9885\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(ytest,por))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "ce95953b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "categories = ['Our method', 'DT', 'RF', 'QBC']\n",
    "values = [0.96, 0.67, 0.67, 0.39]  \n",
    "colors = ['red', 'blue', 'blue', 'blue']\n",
    "fig, ax = plt.subplots()\n",
    "bars = ax.bar(categories, values, color=colors)\n",
    "for bar in bars:\n",
    "    x_val = bar.get_x() + bar.get_width() / 2  \n",
    "    y_val = bar.get_height()  \n",
    "    ax.text(x_val, y_val, f'{y_val:.2f}',  \n",
    "            ha='center', va='bottom')  \n",
    "\n",
    "ax.set_title('Recall per model Top 5 in dataset 3')\n",
    "ax.set_xlabel('Models')\n",
    "ax.set_ylabel('Recall values')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "id": "3b22564f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Enter Gender (0 -> Female, 1 -> Male): \n",
      "Do you own a car? (1 -> own, 0 -> No): \n",
      "What is your reality status? (0 -> No, 1 -> Yes): \n",
      "Number of children: \n",
      "What is your monthly income? (Enter amount): \n",
      "What is your income type? (0 to 4): \n",
      "What is your education level? (0 to 4): \n",
      "What is your family type? (0 to 4): \n",
      "What type of house do you live in? (0 to 5): \n",
      "Do you have a phone? (0 -> No, 1 -> Yes): \n",
      "Do you have an email address? (0 -> No, 1 -> Yes): \n",
      "How many family members are there? \n",
      "In which month did you start? \n",
      "How many years have you been employed? \n",
      "Enter age \n"
     ]
    }
   ],
   "source": [
    "\n",
    "id_input = np.int64(5008899)\n",
    "\n",
    "gender = np.int64(input(\"Enter Gender (0 -> Female, 1 -> Male): \") or 1)\n",
    "car = np.int64(input(\"Do you own a car? (1 -> own, 0 -> No): \") or 0)\n",
    "reality = np.int64(input(\"What is your reality status? (0 -> No, 1 -> Yes): \") or 1)\n",
    "child = np.int64(input(\"Number of children: \") or 1)\n",
    "income = np.int64(input(\"What is your monthly income? (Enter amount): \") or 112500.0)\n",
    "income_type = np.int64(input(\"What is your income type? (0 to 4): \") or 1)\n",
    "education_type = np.int64(input(\"What is your education level? (0 to 4): \") or 1)\n",
    "family_type = np.int64(input(\"What is your family type? (0 to 4): \") or 1)\n",
    "house_type = np.int64(input(\"What type of house do you live in? (0 to 5): \") or 1)\n",
    "work_phone = np.int64(0)\n",
    "phone = np.int64(input(\"Do you have a phone? (0 -> No, 1 -> Yes): \") or 0)\n",
    "email = np.int64(input(\"Do you have an email address? (0 -> No, 1 -> Yes): \") or 0)\n",
    "family_size = np.int64(input(\"How many family members are there? \") or 2)\n",
    "begin_month = np.int64(input(\"In which month did you start? \") or 1)\n",
    "years_employed = np.int64(input(\"How many years have you been employed? \") or 3)\n",
    "age = np.int64(input(\"Enter age \")or 25)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "id": "8d2a40ad",
   "metadata": {},
   "outputs": [],
   "source": [
    "data = {\n",
    "    'Unnamed: 0' :[0],\n",
    "    'ID': [id_input],\n",
    "    'GENDER': [gender],\n",
    "    'CAR': [car],\n",
    "    'REALITY': [reality],\n",
    "    'NO_OF_CHILD': [child],\n",
    "    'INCOME': [income],\n",
    "    'INCOME_TYPE': [income_type],\n",
    "    'EDUCATION_TYPE': [education_type],\n",
    "    'FAMILY_TYPE': [family_type],\n",
    "    'HOUSE_TYPE': [house_type],\n",
    "    'WORK_PHONE': [work_phone],\n",
    "    'PHONE': [phone],\n",
    "    'E_MAIL': [email],\n",
    "    'FAMILY SIZE': [family_size],\n",
    "    'BEGIN_MONTH': [begin_month],\n",
    "    'AGE' : [age],\n",
    "    'YEARS_EMPLOYED': [years_employed],\n",
    "    \n",
    "}\n",
    "\n",
    "df_input = pd.DataFrame(data)\n",
    "ypre_knn_input_new=knn.predict(df_input)\n",
    "ypre_lda_input_new = lda.predict(df_input)\n",
    "ypre_lr_input_new=lr.predict(df_input)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "id": "4eb28e16",
   "metadata": {},
   "outputs": [],
   "source": [
    "ypre_knn_input = (ypre_knn_input_new[0])\n",
    "ypre_lda_input = (ypre_lda_input_new[0]) \n",
    "ypre_lr_input = ypre_lr_input_new[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "id": "5ed14e2f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Legit Transaction \n"
     ]
    }
   ],
   "source": [
    "por = 0 \n",
    "if(ypre_knn_input ==0 or ypre_lda_input==0):\n",
    "    if(ypre_lr_input<0.43):\n",
    "        por=0\n",
    "elif(ypre_input==1 or ypre_lda_input==1):\n",
    "    if(ypre_lr_input>0.43):\n",
    "        por =1\n",
    "else:\n",
    "    por=ypre_knn_input\n",
    "\n",
    "\n",
    "if(por == 1 ):\n",
    "    print(\"TRansaction is Fraud !!!!\")\n",
    "else : \n",
    "    print(\"Legit Transaction \")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "id": "38d0cffe",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Legit Transaction \n"
     ]
    }
   ],
   "source": [
    "dt_pred_new = (clf.predict(df_input)[0])\n",
    "if(dt_pred_new == 1):\n",
    "    print(\"TRansaction is Fraud !!!!\")\n",
    "else : \n",
    "    print(\"Legit Transaction \")"
   ]
  }
 ],
 "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}