RFM_Analysis_Meki_Group.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "2dcbc116",
   "metadata": {},
   "source": [
    "# **RFM (Recency Frequency Monetary) Analysis**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6668ff9f",
   "metadata": {},
   "source": [
    "One of the key strategies for Meki Group in the near future is to improve and enhance the shopping experience for customers, increase the number of new customers, and boost customer retention rates. Therefore, the company wants you to use your analytical skills to calculate and segment new customers and loyal customers based on their purchase history in the provided dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3b2358e0",
   "metadata": {},
   "source": [
    " To segment customers based on their purchase history, we can use the RFM (Recency, Frequency, Monetary) analysis. RFM is a behavioral-based method that categorizes customers into segments. Let’s break down the criteria:\n",
    "\n",
    "- Recency (R): This represents the time since a customer’s most recent purchase. Customers with a smaller recency value tend to be more active and loyal.\n",
    "- Frequency (F): This measures how often a customer makes a purchase within a specific time frame. High-frequency customers are potential repeat buyers.\n",
    "- Monetary (M): This reflects the total amount of money a customer has spent on products during a given period. High monetary value indicates important customers."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8ac6908d",
   "metadata": {},
   "source": [
    "###  Read file\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "7b13c92f",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from datetime import date, datetime\n",
    "import datetime as dt\n",
    "import warnings\n",
    "from sklearn.cluster import KMeans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "9eaf3904",
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "import plotly.express as px\n",
    "import plotly.io as pio\n",
    "import plotly.graph_objects as go\n",
    "pio.templates.default = \"plotly_white\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "1b9fc5f9",
   "metadata": {},
   "outputs": [],
   "source": [
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "0026664d",
   "metadata": {},
   "outputs": [],
   "source": [
    "Sales2020 = pd.read_csv('Historical_sales2020.csv')\n",
    "Sales2021 = pd.read_csv('Historical_sales2021.csv')\n",
    "Sales2022 = pd.read_csv('Historical_sales2022.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "020198a6",
   "metadata": {},
   "source": [
    "### Data cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "bc5fd59e",
   "metadata": {},
   "outputs": [],
   "source": [
    "Sales2021['unit_price'] = Sales2021['unit_price'].str.replace(',','.').astype(float)\n",
    "Sales2020['unit_price'] = Sales2020['unit_price'].str.replace(',','.').astype(float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "3f71a025",
   "metadata": {
    "scrolled": false
   },
   "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>transaction_id</th>\n",
       "      <th>transaction_date</th>\n",
       "      <th>transaction_time</th>\n",
       "      <th>store_id</th>\n",
       "      <th>staff_id</th>\n",
       "      <th>customer_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>quantity_sold</th>\n",
       "      <th>unit_price</th>\n",
       "      <th>promo_item_yn</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>390</td>\n",
       "      <td>1/1/2022</td>\n",
       "      <td>7:01:00</td>\n",
       "      <td>5</td>\n",
       "      <td>12</td>\n",
       "      <td>5658</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1228</td>\n",
       "      <td>1/1/2022</td>\n",
       "      <td>7:02:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5197</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1644</td>\n",
       "      <td>1/1/2022</td>\n",
       "      <td>7:03:00</td>\n",
       "      <td>5</td>\n",
       "      <td>12</td>\n",
       "      <td>5247</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1338</td>\n",
       "      <td>1/1/2022</td>\n",
       "      <td>7:06:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5465</td>\n",
       "      <td>32</td>\n",
       "      <td>2</td>\n",
       "      <td>3.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2257</td>\n",
       "      <td>1/1/2022</td>\n",
       "      <td>7:06:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5627</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553310</th>\n",
       "      <td>401</td>\n",
       "      <td>12/29/2020</td>\n",
       "      <td>18:06:35</td>\n",
       "      <td>8</td>\n",
       "      <td>15</td>\n",
       "      <td>8600</td>\n",
       "      <td>87</td>\n",
       "      <td>1</td>\n",
       "      <td>2.10</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553311</th>\n",
       "      <td>166</td>\n",
       "      <td>12/31/2020</td>\n",
       "      <td>11:00:02</td>\n",
       "      <td>8</td>\n",
       "      <td>42</td>\n",
       "      <td>8219</td>\n",
       "      <td>72</td>\n",
       "      <td>1</td>\n",
       "      <td>3.25</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553312</th>\n",
       "      <td>166</td>\n",
       "      <td>12/31/2020</td>\n",
       "      <td>11:00:02</td>\n",
       "      <td>8</td>\n",
       "      <td>42</td>\n",
       "      <td>8219</td>\n",
       "      <td>87</td>\n",
       "      <td>1</td>\n",
       "      <td>2.10</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553313</th>\n",
       "      <td>296</td>\n",
       "      <td>12/31/2020</td>\n",
       "      <td>15:38:10</td>\n",
       "      <td>5</td>\n",
       "      <td>27</td>\n",
       "      <td>6000</td>\n",
       "      <td>87</td>\n",
       "      <td>1</td>\n",
       "      <td>2.10</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553314</th>\n",
       "      <td>296</td>\n",
       "      <td>12/31/2020</td>\n",
       "      <td>15:38:10</td>\n",
       "      <td>5</td>\n",
       "      <td>27</td>\n",
       "      <td>6000</td>\n",
       "      <td>72</td>\n",
       "      <td>1</td>\n",
       "      <td>2.65</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>553315 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        transaction_id transaction_date transaction_time  store_id  staff_id  \\\n",
       "0                  390         1/1/2022          7:01:00         5        12   \n",
       "1                 1228         1/1/2022          7:02:00         5        30   \n",
       "2                 1644         1/1/2022          7:03:00         5        12   \n",
       "3                 1338         1/1/2022          7:06:00         5        30   \n",
       "4                 2257         1/1/2022          7:06:00         5        30   \n",
       "...                ...              ...              ...       ...       ...   \n",
       "553310             401       12/29/2020         18:06:35         8        15   \n",
       "553311             166       12/31/2020         11:00:02         8        42   \n",
       "553312             166       12/31/2020         11:00:02         8        42   \n",
       "553313             296       12/31/2020         15:38:10         5        27   \n",
       "553314             296       12/31/2020         15:38:10         5        27   \n",
       "\n",
       "        customer_id  product_id  quantity_sold  unit_price promo_item_yn  \n",
       "0              5658          22              2        2.00             N  \n",
       "1              5197          22              1        2.00             N  \n",
       "2              5247          22              2        2.00             N  \n",
       "3              5465          32              2        3.00             N  \n",
       "4              5627          22              1        2.00             N  \n",
       "...             ...         ...            ...         ...           ...  \n",
       "553310         8600          87              1        2.10             Y  \n",
       "553311         8219          72              1        3.25             Y  \n",
       "553312         8219          87              1        2.10             Y  \n",
       "553313         6000          87              1        2.10             Y  \n",
       "553314         6000          72              1        2.65             Y  \n",
       "\n",
       "[553315 rows x 10 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Merge Customer dataset for 3 years\n",
    "Customer_all = pd.concat([Sales2022, Sales2021, Sales2020], ignore_index=True)\n",
    "Customer_all.sort_values('customer_id')\n",
    "Customer_all"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2f5f28f3",
   "metadata": {},
   "outputs": [],
   "source": [
    "Customer_all['transaction_date'] = pd.to_datetime(Customer_all['transaction_date'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ade4882b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "transaction_id      0\n",
       "transaction_date    0\n",
       "transaction_time    0\n",
       "store_id            0\n",
       "staff_id            0\n",
       "customer_id         0\n",
       "product_id          0\n",
       "quantity_sold       0\n",
       "unit_price          0\n",
       "promo_item_yn       0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Customer_all.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "bbad45c3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 553315 entries, 0 to 553314\n",
      "Data columns (total 10 columns):\n",
      " #   Column            Non-Null Count   Dtype         \n",
      "---  ------            --------------   -----         \n",
      " 0   transaction_id    553315 non-null  int64         \n",
      " 1   transaction_date  553315 non-null  datetime64[ns]\n",
      " 2   transaction_time  553315 non-null  object        \n",
      " 3   store_id          553315 non-null  int64         \n",
      " 4   staff_id          553315 non-null  int64         \n",
      " 5   customer_id       553315 non-null  int64         \n",
      " 6   product_id        553315 non-null  int64         \n",
      " 7   quantity_sold     553315 non-null  int64         \n",
      " 8   unit_price        553315 non-null  float64       \n",
      " 9   promo_item_yn     553315 non-null  object        \n",
      "dtypes: datetime64[ns](1), float64(1), int64(6), object(2)\n",
      "memory usage: 42.2+ MB\n"
     ]
    }
   ],
   "source": [
    "Customer_all.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "49479a05",
   "metadata": {
    "scrolled": false
   },
   "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>transaction_id</th>\n",
       "      <th>store_id</th>\n",
       "      <th>staff_id</th>\n",
       "      <th>customer_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>quantity_sold</th>\n",
       "      <th>unit_price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>553315.000000</td>\n",
       "      <td>553315.000000</td>\n",
       "      <td>553315.000000</td>\n",
       "      <td>553315.000000</td>\n",
       "      <td>553315.000000</td>\n",
       "      <td>553315.000000</td>\n",
       "      <td>553315.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>846.754487</td>\n",
       "      <td>5.341393</td>\n",
       "      <td>25.322341</td>\n",
       "      <td>4941.774056</td>\n",
       "      <td>48.072742</td>\n",
       "      <td>1.437331</td>\n",
       "      <td>3.377920</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>849.729672</td>\n",
       "      <td>2.071523</td>\n",
       "      <td>12.441507</td>\n",
       "      <td>3248.569555</td>\n",
       "      <td>18.064032</td>\n",
       "      <td>0.542233</td>\n",
       "      <td>2.656105</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.800000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>216.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>15.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>34.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>457.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>26.000000</td>\n",
       "      <td>5892.000000</td>\n",
       "      <td>47.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1358.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>41.000000</td>\n",
       "      <td>8294.000000</td>\n",
       "      <td>60.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>4203.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>45.000000</td>\n",
       "      <td>8600.000000</td>\n",
       "      <td>87.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>45.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       transaction_id       store_id       staff_id    customer_id  \\\n",
       "count   553315.000000  553315.000000  553315.000000  553315.000000   \n",
       "mean       846.754487       5.341393      25.322341    4941.774056   \n",
       "std        849.729672       2.071523      12.441507    3248.569555   \n",
       "min          1.000000       3.000000       6.000000       1.000000   \n",
       "25%        216.000000       3.000000      15.000000    1000.000000   \n",
       "50%        457.000000       5.000000      26.000000    5892.000000   \n",
       "75%       1358.000000       8.000000      41.000000    8294.000000   \n",
       "max       4203.000000       8.000000      45.000000    8600.000000   \n",
       "\n",
       "          product_id  quantity_sold     unit_price  \n",
       "count  553315.000000  553315.000000  553315.000000  \n",
       "mean       48.072742       1.437331       3.377920  \n",
       "std        18.064032       0.542233       2.656105  \n",
       "min         1.000000       1.000000       0.800000  \n",
       "25%        34.000000       1.000000       2.500000  \n",
       "50%        47.000000       1.000000       3.000000  \n",
       "75%        60.000000       2.000000       3.750000  \n",
       "max        87.000000       8.000000      45.000000  "
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Customer_all.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "06eeda0d",
   "metadata": {},
   "source": [
    "##### Quantity sold and unit price can't less than 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "30811cd9",
   "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>transaction_id</th>\n",
       "      <th>transaction_date</th>\n",
       "      <th>transaction_time</th>\n",
       "      <th>store_id</th>\n",
       "      <th>staff_id</th>\n",
       "      <th>customer_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>quantity_sold</th>\n",
       "      <th>unit_price</th>\n",
       "      <th>promo_item_yn</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [transaction_id, transaction_date, transaction_time, store_id, staff_id, customer_id, product_id, quantity_sold, unit_price, promo_item_yn]\n",
       "Index: []"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Customer_all[Customer_all.quantity_sold < 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "cfa3c7de",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>transaction_id</th>\n",
       "      <th>transaction_date</th>\n",
       "      <th>transaction_time</th>\n",
       "      <th>store_id</th>\n",
       "      <th>staff_id</th>\n",
       "      <th>customer_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>quantity_sold</th>\n",
       "      <th>unit_price</th>\n",
       "      <th>promo_item_yn</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [transaction_id, transaction_date, transaction_time, store_id, staff_id, customer_id, product_id, quantity_sold, unit_price, promo_item_yn]\n",
       "Index: []"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Customer_all[Customer_all.unit_price < 0]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91f8c7d4",
   "metadata": {},
   "source": [
    "#### Remove outlier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "5de538a5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "ax.boxplot(Customer_all['unit_price'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "b56e53b9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.25\n"
     ]
    }
   ],
   "source": [
    "Q1 = np.percentile(Customer_all['unit_price'], 25, method='midpoint')\n",
    "Q3 = np.percentile(Customer_all['unit_price'], 75, method='midpoint')\n",
    "IQR = Q3 - Q1\n",
    "print(IQR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "39542e2c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Upper Bound: 5.625\n",
      "15416\n",
      "Lower Bound: 0.625\n",
      "0\n"
     ]
    }
   ],
   "source": [
    "upper = Q3+1.5*IQR\n",
    "upper_array = np.array(Customer_all['unit_price'] >= upper)\n",
    "print(\"Upper Bound:\", upper)\n",
    "print(upper_array.sum())\n",
    " \n",
    "lower = Q1-1.5*IQR\n",
    "lower_array = np.array(Customer_all['unit_price'] <= lower)\n",
    "print(\"Lower Bound:\", lower)\n",
    "print(lower_array.sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "6ac782c1",
   "metadata": {
    "scrolled": false
   },
   "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>transaction_id</th>\n",
       "      <th>transaction_date</th>\n",
       "      <th>transaction_time</th>\n",
       "      <th>store_id</th>\n",
       "      <th>staff_id</th>\n",
       "      <th>customer_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>quantity_sold</th>\n",
       "      <th>unit_price</th>\n",
       "      <th>promo_item_yn</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>390</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:01:00</td>\n",
       "      <td>5</td>\n",
       "      <td>12</td>\n",
       "      <td>5658</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1228</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:02:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5197</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1644</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:03:00</td>\n",
       "      <td>5</td>\n",
       "      <td>12</td>\n",
       "      <td>5247</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1338</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:06:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5465</td>\n",
       "      <td>32</td>\n",
       "      <td>2</td>\n",
       "      <td>3.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2257</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:06:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5627</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553310</th>\n",
       "      <td>401</td>\n",
       "      <td>2020-12-29</td>\n",
       "      <td>18:06:35</td>\n",
       "      <td>8</td>\n",
       "      <td>15</td>\n",
       "      <td>8600</td>\n",
       "      <td>87</td>\n",
       "      <td>1</td>\n",
       "      <td>2.10</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553311</th>\n",
       "      <td>166</td>\n",
       "      <td>2020-12-31</td>\n",
       "      <td>11:00:02</td>\n",
       "      <td>8</td>\n",
       "      <td>42</td>\n",
       "      <td>8219</td>\n",
       "      <td>72</td>\n",
       "      <td>1</td>\n",
       "      <td>3.25</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553312</th>\n",
       "      <td>166</td>\n",
       "      <td>2020-12-31</td>\n",
       "      <td>11:00:02</td>\n",
       "      <td>8</td>\n",
       "      <td>42</td>\n",
       "      <td>8219</td>\n",
       "      <td>87</td>\n",
       "      <td>1</td>\n",
       "      <td>2.10</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553313</th>\n",
       "      <td>296</td>\n",
       "      <td>2020-12-31</td>\n",
       "      <td>15:38:10</td>\n",
       "      <td>5</td>\n",
       "      <td>27</td>\n",
       "      <td>6000</td>\n",
       "      <td>87</td>\n",
       "      <td>1</td>\n",
       "      <td>2.10</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553314</th>\n",
       "      <td>296</td>\n",
       "      <td>2020-12-31</td>\n",
       "      <td>15:38:10</td>\n",
       "      <td>5</td>\n",
       "      <td>27</td>\n",
       "      <td>6000</td>\n",
       "      <td>72</td>\n",
       "      <td>1</td>\n",
       "      <td>2.65</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>537899 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        transaction_id transaction_date transaction_time  store_id  staff_id  \\\n",
       "0                  390       2022-01-01          7:01:00         5        12   \n",
       "1                 1228       2022-01-01          7:02:00         5        30   \n",
       "2                 1644       2022-01-01          7:03:00         5        12   \n",
       "3                 1338       2022-01-01          7:06:00         5        30   \n",
       "4                 2257       2022-01-01          7:06:00         5        30   \n",
       "...                ...              ...              ...       ...       ...   \n",
       "553310             401       2020-12-29         18:06:35         8        15   \n",
       "553311             166       2020-12-31         11:00:02         8        42   \n",
       "553312             166       2020-12-31         11:00:02         8        42   \n",
       "553313             296       2020-12-31         15:38:10         5        27   \n",
       "553314             296       2020-12-31         15:38:10         5        27   \n",
       "\n",
       "        customer_id  product_id  quantity_sold  unit_price promo_item_yn  \n",
       "0              5658          22              2        2.00             N  \n",
       "1              5197          22              1        2.00             N  \n",
       "2              5247          22              2        2.00             N  \n",
       "3              5465          32              2        3.00             N  \n",
       "4              5627          22              1        2.00             N  \n",
       "...             ...         ...            ...         ...           ...  \n",
       "553310         8600          87              1        2.10             Y  \n",
       "553311         8219          72              1        3.25             Y  \n",
       "553312         8219          87              1        2.10             Y  \n",
       "553313         6000          87              1        2.10             Y  \n",
       "553314         6000          72              1        2.65             Y  \n",
       "\n",
       "[537899 rows x 10 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Customer_all = Customer_all[(Customer_all[\"unit_price\"] > lower) & (Customer_all[\"unit_price\"] < upper)]\n",
    "Customer_all"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "fdda25c5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "ax.boxplot(Customer_all['quantity_sold'])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "a8530f34",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.0\n"
     ]
    }
   ],
   "source": [
    "Q1 = np.percentile(Customer_all['quantity_sold'], 25, method='midpoint')\n",
    "Q3 = np.percentile(Customer_all['quantity_sold'], 75, method='midpoint')\n",
    "IQR = Q3 - Q1\n",
    "print(IQR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "a87639d9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Upper Bound: 3.5\n",
      "92\n",
      "Lower Bound: -0.5\n",
      "0\n"
     ]
    }
   ],
   "source": [
    "upper = Q3+1.5*IQR\n",
    "upper_array = np.array(Customer_all['quantity_sold'] >= upper)\n",
    "print(\"Upper Bound:\", upper)\n",
    "print(upper_array.sum())\n",
    " \n",
    "lower = Q1-1.5*IQR\n",
    "lower_array = np.array(Customer_all['quantity_sold'] <= lower)\n",
    "print(\"Lower Bound:\", lower)\n",
    "print(lower_array.sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "93b18115",
   "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>transaction_id</th>\n",
       "      <th>transaction_date</th>\n",
       "      <th>transaction_time</th>\n",
       "      <th>store_id</th>\n",
       "      <th>staff_id</th>\n",
       "      <th>customer_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>quantity_sold</th>\n",
       "      <th>unit_price</th>\n",
       "      <th>promo_item_yn</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>390</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:01:00</td>\n",
       "      <td>5</td>\n",
       "      <td>12</td>\n",
       "      <td>5658</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1228</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:02:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5197</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1644</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:03:00</td>\n",
       "      <td>5</td>\n",
       "      <td>12</td>\n",
       "      <td>5247</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1338</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:06:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5465</td>\n",
       "      <td>32</td>\n",
       "      <td>2</td>\n",
       "      <td>3.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2257</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:06:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5627</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553310</th>\n",
       "      <td>401</td>\n",
       "      <td>2020-12-29</td>\n",
       "      <td>18:06:35</td>\n",
       "      <td>8</td>\n",
       "      <td>15</td>\n",
       "      <td>8600</td>\n",
       "      <td>87</td>\n",
       "      <td>1</td>\n",
       "      <td>2.10</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553311</th>\n",
       "      <td>166</td>\n",
       "      <td>2020-12-31</td>\n",
       "      <td>11:00:02</td>\n",
       "      <td>8</td>\n",
       "      <td>42</td>\n",
       "      <td>8219</td>\n",
       "      <td>72</td>\n",
       "      <td>1</td>\n",
       "      <td>3.25</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553312</th>\n",
       "      <td>166</td>\n",
       "      <td>2020-12-31</td>\n",
       "      <td>11:00:02</td>\n",
       "      <td>8</td>\n",
       "      <td>42</td>\n",
       "      <td>8219</td>\n",
       "      <td>87</td>\n",
       "      <td>1</td>\n",
       "      <td>2.10</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553313</th>\n",
       "      <td>296</td>\n",
       "      <td>2020-12-31</td>\n",
       "      <td>15:38:10</td>\n",
       "      <td>5</td>\n",
       "      <td>27</td>\n",
       "      <td>6000</td>\n",
       "      <td>87</td>\n",
       "      <td>1</td>\n",
       "      <td>2.10</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>553314</th>\n",
       "      <td>296</td>\n",
       "      <td>2020-12-31</td>\n",
       "      <td>15:38:10</td>\n",
       "      <td>5</td>\n",
       "      <td>27</td>\n",
       "      <td>6000</td>\n",
       "      <td>72</td>\n",
       "      <td>1</td>\n",
       "      <td>2.65</td>\n",
       "      <td>Y</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>537807 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        transaction_id transaction_date transaction_time  store_id  staff_id  \\\n",
       "0                  390       2022-01-01          7:01:00         5        12   \n",
       "1                 1228       2022-01-01          7:02:00         5        30   \n",
       "2                 1644       2022-01-01          7:03:00         5        12   \n",
       "3                 1338       2022-01-01          7:06:00         5        30   \n",
       "4                 2257       2022-01-01          7:06:00         5        30   \n",
       "...                ...              ...              ...       ...       ...   \n",
       "553310             401       2020-12-29         18:06:35         8        15   \n",
       "553311             166       2020-12-31         11:00:02         8        42   \n",
       "553312             166       2020-12-31         11:00:02         8        42   \n",
       "553313             296       2020-12-31         15:38:10         5        27   \n",
       "553314             296       2020-12-31         15:38:10         5        27   \n",
       "\n",
       "        customer_id  product_id  quantity_sold  unit_price promo_item_yn  \n",
       "0              5658          22              2        2.00             N  \n",
       "1              5197          22              1        2.00             N  \n",
       "2              5247          22              2        2.00             N  \n",
       "3              5465          32              2        3.00             N  \n",
       "4              5627          22              1        2.00             N  \n",
       "...             ...         ...            ...         ...           ...  \n",
       "553310         8600          87              1        2.10             Y  \n",
       "553311         8219          72              1        3.25             Y  \n",
       "553312         8219          87              1        2.10             Y  \n",
       "553313         6000          87              1        2.10             Y  \n",
       "553314         6000          72              1        2.65             Y  \n",
       "\n",
       "[537807 rows x 10 columns]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Customer_all = Customer_all[(Customer_all[\"quantity_sold\"] > lower) & (Customer_all[\"quantity_sold\"] < upper)]\n",
    "Customer_all"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01d94a94",
   "metadata": {},
   "source": [
    "## Recency"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "7ffb2b1e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Timestamp('2022-06-28 00:00:00')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "now = dt.date(2022,6,28)\n",
    "now = pd.to_datetime(now)\n",
    "now"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "db53b0fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Last time the customer Purchase \n",
    "Recency = Customer_all.groupby(by='customer_id',as_index=False)['transaction_date'].max()\n",
    "Recency.columns = ['Customer_id', 'Last_Purchase_Date']\n",
    "\n",
    "# Convert Last_Purchase_Date to datetime\n",
    "Recency[\"Last_Purchase_Date\"] = pd.to_datetime(Recency[\"Last_Purchase_Date\"])\n",
    "\n",
    "#Calculate Recency\n",
    "Recency['Recency'] = Recency.Last_Purchase_Date.apply(lambda x : (now - x).days)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "fad0b84f",
   "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>Customer_id</th>\n",
       "      <th>Last_Purchase_Date</th>\n",
       "      <th>Recency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>2022-04-29</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>2022-04-26</td>\n",
       "      <td>63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>2022-04-24</td>\n",
       "      <td>65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>2022-04-26</td>\n",
       "      <td>63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>2022-04-29</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2245</th>\n",
       "      <td>8498</td>\n",
       "      <td>2022-04-27</td>\n",
       "      <td>62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2246</th>\n",
       "      <td>8499</td>\n",
       "      <td>2022-04-29</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2247</th>\n",
       "      <td>8500</td>\n",
       "      <td>2022-04-26</td>\n",
       "      <td>63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2248</th>\n",
       "      <td>8501</td>\n",
       "      <td>2022-04-28</td>\n",
       "      <td>61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2249</th>\n",
       "      <td>8600</td>\n",
       "      <td>2022-04-29</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2250 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Customer_id Last_Purchase_Date  Recency\n",
       "0               1         2022-04-29       60\n",
       "1               2         2022-04-26       63\n",
       "2               3         2022-04-24       65\n",
       "3               4         2022-04-26       63\n",
       "4               5         2022-04-29       60\n",
       "...           ...                ...      ...\n",
       "2245         8498         2022-04-27       62\n",
       "2246         8499         2022-04-29       60\n",
       "2247         8500         2022-04-26       63\n",
       "2248         8501         2022-04-28       61\n",
       "2249         8600         2022-04-29       60\n",
       "\n",
       "[2250 rows x 3 columns]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Recency"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "f2f8ee2e",
   "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>Customer_id</th>\n",
       "      <th>Recency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>60</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Customer_id  Recency\n",
       "0            1       60\n",
       "1            2       63\n",
       "2            3       65\n",
       "3            4       63\n",
       "4            5       60"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Recency.drop(columns=['Last_Purchase_Date'],inplace=True)\n",
    "Recency.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "42a188b3",
   "metadata": {},
   "source": [
    "## Frequency"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "9e4137a5",
   "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>transaction_id</th>\n",
       "      <th>transaction_date</th>\n",
       "      <th>transaction_time</th>\n",
       "      <th>store_id</th>\n",
       "      <th>staff_id</th>\n",
       "      <th>customer_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>quantity_sold</th>\n",
       "      <th>unit_price</th>\n",
       "      <th>promo_item_yn</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>390</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:01:00</td>\n",
       "      <td>5</td>\n",
       "      <td>12</td>\n",
       "      <td>5658</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1228</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:02:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5197</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1644</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:03:00</td>\n",
       "      <td>5</td>\n",
       "      <td>12</td>\n",
       "      <td>5247</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1338</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:06:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5465</td>\n",
       "      <td>32</td>\n",
       "      <td>2</td>\n",
       "      <td>3.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2257</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:06:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5627</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>2.00</td>\n",
       "      <td>N</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>156958</th>\n",
       "      <td>1950</td>\n",
       "      <td>2021-01-17</td>\n",
       "      <td>15:50</td>\n",
       "      <td>8</td>\n",
       "      <td>43</td>\n",
       "      <td>8482</td>\n",
       "      <td>31</td>\n",
       "      <td>1</td>\n",
       "      <td>2.20</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>186090</th>\n",
       "      <td>547</td>\n",
       "      <td>2021-02-17</td>\n",
       "      <td>10:00</td>\n",
       "      <td>8</td>\n",
       "      <td>44</td>\n",
       "      <td>8114</td>\n",
       "      <td>40</td>\n",
       "      <td>2</td>\n",
       "      <td>3.75</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>213873</th>\n",
       "      <td>1633</td>\n",
       "      <td>2021-03-17</td>\n",
       "      <td>11:57</td>\n",
       "      <td>8</td>\n",
       "      <td>43</td>\n",
       "      <td>8074</td>\n",
       "      <td>60</td>\n",
       "      <td>2</td>\n",
       "      <td>3.75</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>245217</th>\n",
       "      <td>1365</td>\n",
       "      <td>2021-04-17</td>\n",
       "      <td>7:16</td>\n",
       "      <td>8</td>\n",
       "      <td>44</td>\n",
       "      <td>8405</td>\n",
       "      <td>40</td>\n",
       "      <td>2</td>\n",
       "      <td>3.75</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>245685</th>\n",
       "      <td>1474</td>\n",
       "      <td>2021-04-17</td>\n",
       "      <td>10:31</td>\n",
       "      <td>8</td>\n",
       "      <td>6</td>\n",
       "      <td>8166</td>\n",
       "      <td>24</td>\n",
       "      <td>2</td>\n",
       "      <td>3.00</td>\n",
       "      <td>N</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>27257 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        transaction_id transaction_date transaction_time  store_id  staff_id  \\\n",
       "0                  390       2022-01-01          7:01:00         5        12   \n",
       "1                 1228       2022-01-01          7:02:00         5        30   \n",
       "2                 1644       2022-01-01          7:03:00         5        12   \n",
       "3                 1338       2022-01-01          7:06:00         5        30   \n",
       "4                 2257       2022-01-01          7:06:00         5        30   \n",
       "...                ...              ...              ...       ...       ...   \n",
       "156958            1950       2021-01-17            15:50         8        43   \n",
       "186090             547       2021-02-17            10:00         8        44   \n",
       "213873            1633       2021-03-17            11:57         8        43   \n",
       "245217            1365       2021-04-17             7:16         8        44   \n",
       "245685            1474       2021-04-17            10:31         8         6   \n",
       "\n",
       "        customer_id  product_id  quantity_sold  unit_price promo_item_yn  \n",
       "0              5658          22              2        2.00             N  \n",
       "1              5197          22              1        2.00             N  \n",
       "2              5247          22              2        2.00             N  \n",
       "3              5465          32              2        3.00             N  \n",
       "4              5627          22              1        2.00             N  \n",
       "...             ...         ...            ...         ...           ...  \n",
       "156958         8482          31              1        2.20             N  \n",
       "186090         8114          40              2        3.75             N  \n",
       "213873         8074          60              2        3.75             N  \n",
       "245217         8405          40              2        3.75             N  \n",
       "245685         8166          24              2        3.00             N  \n",
       "\n",
       "[27257 rows x 10 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Frequency = Customer_all.copy()\n",
    "Frequency.drop_duplicates(subset=['customer_id','transaction_id'],keep=\"first\",inplace=True)\n",
    "Frequency"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "06b71338",
   "metadata": {},
   "outputs": [],
   "source": [
    "Frequency = Frequency.drop_duplicates().groupby(\n",
    "    by=['customer_id'], as_index=False)['transaction_date'].count()\n",
    "Frequency.columns = ['Customer_id', 'Frequency']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "5e3f5558",
   "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>Customer_id</th>\n",
       "      <th>Frequency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2245</th>\n",
       "      <td>8498</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2246</th>\n",
       "      <td>8499</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2247</th>\n",
       "      <td>8500</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2248</th>\n",
       "      <td>8501</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2249</th>\n",
       "      <td>8600</td>\n",
       "      <td>2258</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2250 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Customer_id  Frequency\n",
       "0               1          6\n",
       "1               2         16\n",
       "2               3         13\n",
       "3               4          7\n",
       "4               5          6\n",
       "...           ...        ...\n",
       "2245         8498          7\n",
       "2246         8499         10\n",
       "2247         8500         17\n",
       "2248         8501          6\n",
       "2249         8600       2258\n",
       "\n",
       "[2250 rows x 2 columns]"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Frequency"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "082e4d5f",
   "metadata": {},
   "source": [
    "## Monetary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "8f25a445",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate the revenue of company \n",
    "Customer_all['Revenue'] = Customer_all['unit_price'] * Customer_all['quantity_sold']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "f2743587",
   "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>transaction_id</th>\n",
       "      <th>transaction_date</th>\n",
       "      <th>transaction_time</th>\n",
       "      <th>store_id</th>\n",
       "      <th>staff_id</th>\n",
       "      <th>customer_id</th>\n",
       "      <th>product_id</th>\n",
       "      <th>quantity_sold</th>\n",
       "      <th>unit_price</th>\n",
       "      <th>promo_item_yn</th>\n",
       "      <th>Revenue</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>390</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:01:00</td>\n",
       "      <td>5</td>\n",
       "      <td>12</td>\n",
       "      <td>5658</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>N</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1228</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:02:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5197</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>N</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1644</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:03:00</td>\n",
       "      <td>5</td>\n",
       "      <td>12</td>\n",
       "      <td>5247</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>N</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1338</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:06:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5465</td>\n",
       "      <td>32</td>\n",
       "      <td>2</td>\n",
       "      <td>3.0</td>\n",
       "      <td>N</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2257</td>\n",
       "      <td>2022-01-01</td>\n",
       "      <td>7:06:00</td>\n",
       "      <td>5</td>\n",
       "      <td>30</td>\n",
       "      <td>5627</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>N</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   transaction_id transaction_date transaction_time  store_id  staff_id  \\\n",
       "0             390       2022-01-01          7:01:00         5        12   \n",
       "1            1228       2022-01-01          7:02:00         5        30   \n",
       "2            1644       2022-01-01          7:03:00         5        12   \n",
       "3            1338       2022-01-01          7:06:00         5        30   \n",
       "4            2257       2022-01-01          7:06:00         5        30   \n",
       "\n",
       "   customer_id  product_id  quantity_sold  unit_price promo_item_yn  Revenue  \n",
       "0         5658          22              2         2.0             N      4.0  \n",
       "1         5197          22              1         2.0             N      2.0  \n",
       "2         5247          22              2         2.0             N      4.0  \n",
       "3         5465          32              2         3.0             N      6.0  \n",
       "4         5627          22              1         2.0             N      2.0  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Customer_all.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "1ebedad6",
   "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>Customer_id</th>\n",
       "      <th>Monetary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>333.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>913.65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>1390.85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>314.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>276.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Customer_id  Monetary\n",
       "0            1    333.00\n",
       "1            2    913.65\n",
       "2            3   1390.85\n",
       "3            4    314.75\n",
       "4            5    276.00"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Find Revenue from each Customer\n",
    "Monetary = Customer_all.groupby(by='customer_id', as_index=False)['Revenue'].sum()\n",
    "Monetary.columns = ['Customer_id', 'Monetary']\n",
    "Monetary.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fa859c37",
   "metadata": {},
   "source": [
    "## Merging all three columns in one dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "d2b53614",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Customer_id</th>\n",
       "      <th>Recency</th>\n",
       "      <th>Frequency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>60</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>63</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>65</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>63</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>60</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Customer_id  Recency  Frequency\n",
       "0            1       60          6\n",
       "1            2       63         16\n",
       "2            3       65         13\n",
       "3            4       63          7\n",
       "4            5       60          6"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Merge Recency with Frequency\n",
    "RF = Recency.merge(Frequency,left_on='Customer_id',right_on='Customer_id')\n",
    "RF.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "9019639b",
   "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>Recency</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Monetary</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Customer_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>60</td>\n",
       "      <td>6</td>\n",
       "      <td>333.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>63</td>\n",
       "      <td>16</td>\n",
       "      <td>913.65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>65</td>\n",
       "      <td>13</td>\n",
       "      <td>1390.85</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>63</td>\n",
       "      <td>7</td>\n",
       "      <td>314.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>60</td>\n",
       "      <td>6</td>\n",
       "      <td>276.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Recency  Frequency  Monetary\n",
       "Customer_id                              \n",
       "1                 60          6    333.00\n",
       "2                 63         16    913.65\n",
       "3                 65         13   1390.85\n",
       "4                 63          7    314.75\n",
       "5                 60          6    276.00"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Merge with Monetary\n",
    "RFM = RF.merge(Monetary, left_on='Customer_id',right_on='Customer_id')\n",
    "RFM.set_index('Customer_id',inplace=True)\n",
    "RFM.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a8ce3b07",
   "metadata": {},
   "source": [
    "## Clustering with K means"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "e247db10",
   "metadata": {},
   "outputs": [],
   "source": [
    "rfm_segmentation = RFM.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "72501d77",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Nc = range(1, 10)\n",
    "kmeans = [KMeans(n_clusters=i) for i in Nc]\n",
    "score = [kmeans[i].fit(rfm_segmentation).score(rfm_segmentation) for i in range(len(kmeans))]\n",
    "plt.plot(Nc,score,marker='*')\n",
    "plt.xlabel('Number of Clusters')\n",
    "plt.ylabel('Score')\n",
    "plt.grid()\n",
    "plt.title('Elbow Curve')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "1c33ead6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Choose number of clusters is K=3\n",
    "kmeans = KMeans(n_clusters = 3, random_state=0).fit(rfm_segmentation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "478c4ca3",
   "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>Recency</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Monetary</th>\n",
       "      <th>cluster</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Customer_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>60</td>\n",
       "      <td>6</td>\n",
       "      <td>333.00</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>63</td>\n",
       "      <td>16</td>\n",
       "      <td>913.65</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>65</td>\n",
       "      <td>13</td>\n",
       "      <td>1390.85</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>63</td>\n",
       "      <td>7</td>\n",
       "      <td>314.75</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>60</td>\n",
       "      <td>6</td>\n",
       "      <td>276.00</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>68</td>\n",
       "      <td>9</td>\n",
       "      <td>511.80</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>61</td>\n",
       "      <td>9</td>\n",
       "      <td>595.10</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>63</td>\n",
       "      <td>10</td>\n",
       "      <td>631.05</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>63</td>\n",
       "      <td>9</td>\n",
       "      <td>522.65</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>60</td>\n",
       "      <td>10</td>\n",
       "      <td>504.20</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Recency  Frequency  Monetary  cluster\n",
       "Customer_id                                       \n",
       "1                 60          6    333.00        0\n",
       "2                 63         16    913.65        0\n",
       "3                 65         13   1390.85        0\n",
       "4                 63          7    314.75        0\n",
       "5                 60          6    276.00        0\n",
       "6                 68          9    511.80        0\n",
       "7                 61          9    595.10        0\n",
       "8                 63         10    631.05        0\n",
       "9                 63          9    522.65        0\n",
       "10                60         10    504.20        0"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# add labels after clustering \n",
    "rfm_segmentation['cluster'] = kmeans.labels_\n",
    "rfm_segmentation.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "751eb7b9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    2247\n",
       "1       2\n",
       "2       1\n",
       "Name: cluster, dtype: int64"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Quantity each cluster\n",
    "rfm_segmentation['cluster'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "8ef69e29",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='cluster', ylabel='Monetary'>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(x='cluster', y='Monetary', data=rfm_segmentation)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "70aefe0c",
   "metadata": {},
   "source": [
    "- cluster 0 have low Monetary rate which is bad. cluster 2 have highest Montary (money spend) platinum where as cluster 1 have medium level(Gold) and cluster 0 is silver customer. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "5d1db192",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='cluster', ylabel='Recency'>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(x='cluster', y='Recency', data=rfm_segmentation)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d9222e36",
   "metadata": {},
   "source": [
    "- cluster 0 have high recency rate which is bad. cluster 1 and cluster 2 having low so they are in race of platinum and gold customer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "535f0fac",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='cluster', ylabel='Frequency'>"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.boxplot(x='cluster', y='Frequency', data=rfm_segmentation)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "652eee0f",
   "metadata": {},
   "source": [
    "- cluster 0 have low frequency rate which is bad. cluster 1 and cluster 2 having high so they are in race of platinum and gold customer."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "72e90dcc",
   "metadata": {},
   "source": [
    "### **Conclusion**\n",
    "---\n",
    "1. Platinum customer - frequent and more revenue generator.\n",
    "2. Gold customer - frequent but less revenue generator.\n",
    "3. silver customer - less frequent and less revenue generator."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fcb49923",
   "metadata": {},
   "source": [
    "## Ranking Customer’s based upon their recency, frequency, and monetary score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "6350e552",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Ranking fine scoring criteria for each RFM value\n",
    "# Arguments (x = value, p = recency, monetary_value, frequency, d = quartiles dict)\n",
    "\n",
    "#lower the recency, good for store..\n",
    "def RScore(x,p,d):\n",
    "    if x <= d[p][0.25]:\n",
    "        return 1\n",
    "    elif x <= d[p][0.50]:\n",
    "        return 2\n",
    "    elif x <= d[p][0.75]:\n",
    "        return 3\n",
    "    else:\n",
    "        return 4\n",
    "\n",
    "\n",
    "# Arguments (x = value, p = recency, monetary_value, frequency, k = quartiles dict)\n",
    "\n",
    "#higher value of frequency and monetary lead to a good consumer. here higher value = 1 in reverse way.\n",
    "\n",
    "def FMScore(x,p,d):\n",
    "    if x <= d[p][0.25]:\n",
    "        return 4\n",
    "    elif x <= d[p][0.50]:\n",
    "        return 3\n",
    "    elif x <= d[p][0.75]:\n",
    "        return 2\n",
    "    else:\n",
    "        return 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "33ac387e",
   "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>Recency</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Monetary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.25</th>\n",
       "      <td>61.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>353.3125</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.50</th>\n",
       "      <td>62.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>498.4500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.75</th>\n",
       "      <td>65.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>670.0125</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Recency  Frequency  Monetary\n",
       "0.25     61.0        6.0  353.3125\n",
       "0.50     62.0        8.0  498.4500\n",
       "0.75     65.0       11.0  670.0125"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculate RFM scores\n",
    "quantile = RFM.quantile(q=[0.25,0.5,0.75])\n",
    "quantile"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "333ac1ab",
   "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>Recency</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Monetary</th>\n",
       "      <th>cluster</th>\n",
       "      <th>R_Quartile</th>\n",
       "      <th>F_Quartile</th>\n",
       "      <th>M_Quartile</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Customer_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>60</td>\n",
       "      <td>6</td>\n",
       "      <td>333.00</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>63</td>\n",
       "      <td>16</td>\n",
       "      <td>913.65</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>65</td>\n",
       "      <td>13</td>\n",
       "      <td>1390.85</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>63</td>\n",
       "      <td>7</td>\n",
       "      <td>314.75</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>60</td>\n",
       "      <td>6</td>\n",
       "      <td>276.00</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Recency  Frequency  Monetary  cluster  R_Quartile  F_Quartile  \\\n",
       "Customer_id                                                                  \n",
       "1                 60          6    333.00        0           1           4   \n",
       "2                 63         16    913.65        0           3           1   \n",
       "3                 65         13   1390.85        0           3           1   \n",
       "4                 63          7    314.75        0           3           3   \n",
       "5                 60          6    276.00        0           1           4   \n",
       "\n",
       "             M_Quartile  \n",
       "Customer_id              \n",
       "1                     4  \n",
       "2                     1  \n",
       "3                     1  \n",
       "4                     4  \n",
       "5                     4  "
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Convert RFM scores to numeric type\n",
    "rfm_segmentation['R_Quartile'] = rfm_segmentation['Recency'].apply(RScore,args=('Recency',quantile))\n",
    "rfm_segmentation['F_Quartile'] = rfm_segmentation['Frequency'].apply(FMScore, args=('Frequency',quantile))\n",
    "rfm_segmentation['M_Quartile'] = rfm_segmentation['Monetary'].apply(FMScore, args=('Monetary',quantile))\n",
    "rfm_segmentation.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "5b9263ec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Recency': {0.25: 61.0, 0.5: 62.0, 0.75: 65.0},\n",
       " 'Frequency': {0.25: 6.0, 0.5: 8.0, 0.75: 11.0},\n",
       " 'Monetary': {0.25: 353.3125, 0.5: 498.45, 0.75: 670.0125}}"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "quantile.to_dict()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "345894b8",
   "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>Recency</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Monetary</th>\n",
       "      <th>cluster</th>\n",
       "      <th>R_Quartile</th>\n",
       "      <th>F_Quartile</th>\n",
       "      <th>M_Quartile</th>\n",
       "      <th>RFMScore</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Customer_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>60</td>\n",
       "      <td>6</td>\n",
       "      <td>333.00</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>63</td>\n",
       "      <td>16</td>\n",
       "      <td>913.65</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>65</td>\n",
       "      <td>13</td>\n",
       "      <td>1390.85</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>311</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>63</td>\n",
       "      <td>7</td>\n",
       "      <td>314.75</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>60</td>\n",
       "      <td>6</td>\n",
       "      <td>276.00</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>144</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Recency  Frequency  Monetary  cluster  R_Quartile  F_Quartile  \\\n",
       "Customer_id                                                                  \n",
       "1                 60          6    333.00        0           1           4   \n",
       "2                 63         16    913.65        0           3           1   \n",
       "3                 65         13   1390.85        0           3           1   \n",
       "4                 63          7    314.75        0           3           3   \n",
       "5                 60          6    276.00        0           1           4   \n",
       "\n",
       "             M_Quartile RFMScore  \n",
       "Customer_id                       \n",
       "1                     4      144  \n",
       "2                     1      311  \n",
       "3                     1      311  \n",
       "4                     4      334  \n",
       "5                     4      144  "
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#calculate RFM score\n",
    "\n",
    "rfm_segmentation['RFMScore'] = rfm_segmentation.R_Quartile.map(str) \\\n",
    "                            + rfm_segmentation.F_Quartile.map(str) \\\n",
    "                            + rfm_segmentation.M_Quartile.map(str)\n",
    "rfm_segmentation.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "d42f3d03",
   "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>Recency</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Monetary</th>\n",
       "      <th>cluster</th>\n",
       "      <th>R_Quartile</th>\n",
       "      <th>F_Quartile</th>\n",
       "      <th>M_Quartile</th>\n",
       "      <th>RFMScore</th>\n",
       "      <th>Total_score</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Customer_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>60</td>\n",
       "      <td>6</td>\n",
       "      <td>333.00</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>144</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>63</td>\n",
       "      <td>16</td>\n",
       "      <td>913.65</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>311</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>65</td>\n",
       "      <td>13</td>\n",
       "      <td>1390.85</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>311</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>63</td>\n",
       "      <td>7</td>\n",
       "      <td>314.75</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>334</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>60</td>\n",
       "      <td>6</td>\n",
       "      <td>276.00</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>144</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Recency  Frequency  Monetary  cluster  R_Quartile  F_Quartile  \\\n",
       "Customer_id                                                                  \n",
       "1                 60          6    333.00        0           1           4   \n",
       "2                 63         16    913.65        0           3           1   \n",
       "3                 65         13   1390.85        0           3           1   \n",
       "4                 63          7    314.75        0           3           3   \n",
       "5                 60          6    276.00        0           1           4   \n",
       "\n",
       "             M_Quartile RFMScore  Total_score  \n",
       "Customer_id                                    \n",
       "1                     4      144            9  \n",
       "2                     1      311            5  \n",
       "3                     1      311            5  \n",
       "4                     4      334           10  \n",
       "5                     4      144            9  "
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfm_segmentation['Total_score'] = rfm_segmentation['R_Quartile'] + \\\n",
    "                                rfm_segmentation['F_Quartile'] + \\\n",
    "                                rfm_segmentation['M_Quartile']\n",
    "\n",
    "rfm_segmentation.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "66916a38",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Customers:  207\n",
      "Loyal Customers:  474\n",
      "Big Spenders:  563\n",
      "Almost Lost:  23\n",
      "Lost Customers:  76\n",
      "Lost Cheap Customers:  222\n"
     ]
    }
   ],
   "source": [
    "print(\"Best Customers: \",len(rfm_segmentation[rfm_segmentation['RFMScore']=='111']))\n",
    "print('Loyal Customers: ',len(rfm_segmentation[rfm_segmentation['F_Quartile']==1]))\n",
    "print(\"Big Spenders: \",len(rfm_segmentation[rfm_segmentation['M_Quartile']==1]))\n",
    "print('Almost Lost: ', len(rfm_segmentation[rfm_segmentation['RFMScore']=='134']))\n",
    "print('Lost Customers: ',len(rfm_segmentation[rfm_segmentation['RFMScore']=='344']))\n",
    "print('Lost Cheap Customers: ',len(rfm_segmentation[rfm_segmentation['RFMScore']=='444']))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "6171612b",
   "metadata": {},
   "outputs": [],
   "source": [
    "categories = ['Best Customers', 'Loyal Customers', 'Big Spenders', 'Almost Lost', 'Lost Customers', 'Lost Cheap Customers']\n",
    "quantities = [207, 474, 563, 23, 76, 222]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "18d433e4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "colors = ['#FF9999', '#66B3FF', '#99FF99', '#FFCC99', '#FFD700', '#C71585']\n",
    "\n",
    "# Create the bar chart\n",
    "plt.figure(figsize=(10, 6))\n",
    "bars = plt.bar(categories, quantities, color= colors)\n",
    "\n",
    "# Add titles and labels\n",
    "plt.title('Customer Segmentation')\n",
    "plt.xlabel('Customer Segments')\n",
    "plt.ylabel('Quantity')\n",
    "plt.xticks(rotation=45)\n",
    "\n",
    "for bar in bars:\n",
    "    yval = bar.get_height()\n",
    "    plt.text(bar.get_x() + bar.get_width()/2.0, yval, int(yval), va='bottom')\n",
    "    \n",
    "# Display the bar chart\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2f100b1",
   "metadata": {},
   "source": [
    "### Some strategic development directions for the store:\n",
    "\n",
    "__Enhance Loyal Customer Care:__\n",
    "\n",
    "- Reward Programs: Develop reward programs, discounts, or special offers exclusively for loyal customers to increase engagement.\n",
    "- Personalized Services: Provide personalized services, such as sending birthday wishes or special gifts on significant occasions.\n",
    "\n",
    "__Attract High-Spending Customers:__\n",
    "\n",
    "- Exclusive Offers: Provide exclusive offers and VIP customer programs for high-spending groups.\n",
    "- Premium Products/Services: Introduce additional premium products or services to meet the needs of this customer group.\n",
    "\n",
    "__Recover Nearly Lost Customers:__\n",
    "\n",
    "- Feedback Surveys: Send feedback surveys to understand why customers almost left and improve weak points.\n",
    "- Promotional Campaigns: Implement promotional campaigns aimed at enticing them to return, such as discounts or gifts when they shop again.\n",
    "\n",
    "__Reduce Customer Attrition:__\n",
    "\n",
    "- Analyze Reasons: Analyze the reasons for customer attrition and take measures to address them, such as improving product or service quality.\n",
    "- Promotional Programs: Create attractive promotional programs to retain customers.\n",
    "\n",
    "__Optimize Best Customer Care:__\n",
    "\n",
    "- Priority Services: Provide priority and special services for the best customer group to enhance satisfaction and loyalty.\n",
    "- Loyalty Programs: Ensure this customer group always benefits from loyalty programs and feels valued. \n",
    "\n",
    "__Marketing Strategy for Potential Customers:__\n",
    "\n",
    "- Marketing Campaigns: Build marketing campaigns targeting potential customers based on the profile of current customers.\n",
    "- Targeted Advertising: Use targeted advertising on online platforms to reach potential customer groups.\n",
    "\n",
    "__Improve Overall Customer Experience:__\n",
    "\n",
    "- Employee Training: Enhance employee training on customer service to ensure every customer has the best experience when visiting the store.\n",
    "- Service Improvement: Continuously improve service and product quality based on customer feedback."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "9dcff5c6",
   "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>Recency</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>Monetary</th>\n",
       "      <th>cluster</th>\n",
       "      <th>R_Quartile</th>\n",
       "      <th>F_Quartile</th>\n",
       "      <th>M_Quartile</th>\n",
       "      <th>RFMScore</th>\n",
       "      <th>Total_score</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Customer_id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>8600</th>\n",
       "      <td>60</td>\n",
       "      <td>2258</td>\n",
       "      <td>416881.00</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>111</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1000</th>\n",
       "      <td>60</td>\n",
       "      <td>3444</td>\n",
       "      <td>379017.45</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>111</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8285</th>\n",
       "      <td>60</td>\n",
       "      <td>25</td>\n",
       "      <td>1564.95</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>111</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8297</th>\n",
       "      <td>60</td>\n",
       "      <td>23</td>\n",
       "      <td>1495.15</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>111</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8384</th>\n",
       "      <td>60</td>\n",
       "      <td>23</td>\n",
       "      <td>1488.90</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>111</td>\n",
       "      <td>3</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>70</td>\n",
       "      <td>1</td>\n",
       "      <td>65.10</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>444</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5813</th>\n",
       "      <td>78</td>\n",
       "      <td>2</td>\n",
       "      <td>64.00</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>444</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>359</th>\n",
       "      <td>86</td>\n",
       "      <td>1</td>\n",
       "      <td>55.00</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>444</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5147</th>\n",
       "      <td>84</td>\n",
       "      <td>1</td>\n",
       "      <td>36.00</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>444</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>427</th>\n",
       "      <td>85</td>\n",
       "      <td>1</td>\n",
       "      <td>28.00</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>444</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2250 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             Recency  Frequency   Monetary  cluster  R_Quartile  F_Quartile  \\\n",
       "Customer_id                                                                   \n",
       "8600              60       2258  416881.00        2           1           1   \n",
       "1000              60       3444  379017.45        1           1           1   \n",
       "8285              60         25    1564.95        0           1           1   \n",
       "8297              60         23    1495.15        0           1           1   \n",
       "8384              60         23    1488.90        0           1           1   \n",
       "...              ...        ...        ...      ...         ...         ...   \n",
       "17                70          1      65.10        0           4           4   \n",
       "5813              78          2      64.00        0           4           4   \n",
       "359               86          1      55.00        0           4           4   \n",
       "5147              84          1      36.00        0           4           4   \n",
       "427               85          1      28.00        0           4           4   \n",
       "\n",
       "             M_Quartile RFMScore  Total_score  \n",
       "Customer_id                                    \n",
       "8600                  1      111            3  \n",
       "1000                  1      111            3  \n",
       "8285                  1      111            3  \n",
       "8297                  1      111            3  \n",
       "8384                  1      111            3  \n",
       "...                 ...      ...          ...  \n",
       "17                    4      444           12  \n",
       "5813                  4      444           12  \n",
       "359                   4      444           12  \n",
       "5147                  4      444           12  \n",
       "427                   4      444           12  \n",
       "\n",
       "[2250 rows x 9 columns]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfm_segmentation.sort_values(by=['RFMScore', 'Monetary'], ascending=[True, False])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "af3310be",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Average money value of each RFM group')"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "rfm_segmentation.groupby('Total_score')['Monetary'].mean().plot(kind='bar', colormap='viridis_r')\n",
    "\n",
    "plt.title('Average money value of each RFM group')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6d27085d",
   "metadata": {},
   "source": [
    "__Based on Monetary we see Total_score 3,4,5 have highest value which is good for model. because it could have combination of values such as 111, 121, 122__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "b2924f31",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Average Frequency value of each RFM group')"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "rfm_segmentation.groupby('Total_score')['Frequency'].mean().plot(kind='bar', colormap='coolwarm')\n",
    "\n",
    "plt.title('Average Frequency value of each RFM group')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bc9fae4e",
   "metadata": {},
   "source": [
    "__Based on Frequncy we see Total_score 3,4,5 have highest value which is good for model. because it could have\n",
    "combination of values such as 111, 121, 122__"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "54f11282",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Average Recency value of each RFM group')"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAh8AAAHJCAYAAAAl0lt+AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAABBmElEQVR4nO3df3zOdf////sx49gxtvm9GTPDEiHCyaba+mF9VCLhjc4i59mpVJJKp5wV1blJZ1IpIfnZSj9UlF+TUmc4jTPSKumkrJOhYlvSxvb4/uG743TYsGPmNVu36+Xyulwcz9frdbwej2PHjuPuebxex1xmZgIAAHBIQEUXAAAAfl8IHwAAwFGEDwAA4CjCBwAAcBThAwAAOIrwAQAAHEX4AAAAjiJ8AAAARxE+AACAowgfldyzzz4rl8ultm3bVnQp55zExES5XC7vEhQUpDZt2ujxxx9Xfn5+RZdXpc2ZM0cul0vfffddRZdySs8995xatmypGjVqyOVy6eDBgxVdko9mzZrp2muvLfP+xz//XS6XQkNDFR8fr1dffbXYtkU/s5KW++67z6cml8ulxMTEEo85b948734fffRRmWtH1RZY0QXgzLz88suSpIyMDP3rX/9S165dK7iic0vz5s31yiuvSJL279+vl156SQ899JB27dqlGTNmVHB1qEibN2/WyJEj9ec//1lDhgxRYGCgQkJCKrqsctevXz/de++9MjPt3LlTycnJGjx4sMxMgwcPLrb97Nmzdf755/uMRUZG+twOCQnRxx9/rP/85z9q0aKFz7qXX35ZoaGhysnJKf9mUGUw81GJbdy4UVu2bNE111wjSZo1a5bjNZiZDh8+7PhxS8vj8ahbt27q1q2bevXqpbfeekuxsbGaO3eufvvtt4ouDxUoIyNDknTrrbfq4osvVrdu3VStWrUKrqr8hYeHq1u3boqLi9PgwYP1/vvvS5KmT59e4vZt27b1/s4ULU2bNvXZ5uKLL1bjxo29//kp8p///Ecff/yx/u///u/sNHOCc/m1B6dG+KjEisLGxIkTFR8fr9dee02//vqrJOnIkSNq2LChbrrppmL7HTx4UB6PR6NHj/aO5eTk6L777lNMTIxq1Kihxo0ba9SoUTp06JDPvi6XS3feeadefPFFtW7dWm63W3PnzpUkTZgwQV27dlXdunUVGhqqiy66SLNmzdKJf7swLy9P9957ryIiIhQcHKxLL71UmzZtUrNmzTR06FCfbbOysjR8+HA1adJENWrUUExMjCZMmKCjR4+W6TELDAxUhw4dlJ+f7zPFbmZ64YUX1KFDB3k8HtWpU0f9+vXTjh07it3H8uXLdcUVVygsLEzBwcFq3bq1UlJSfLbZuHGjrrvuOtWtW1dBQUHq2LGjXn/9dZ9tiqa5P/zwQ91+++2qX7++6tWrp759+2r37t3Fjpuamqq4uDjVqlVLtWrVUocOHbzPgccee0yBgYHKzMwstt+wYcNUr169k4atKVOmyOVy6dtvvy227oEHHlCNGjX0448/SpLS0tLUu3dvNWnSREFBQWrZsqWGDx/uXX8qJf18pWMfj504hV/a5+PJvPzyy7rwwgsVFBSkunXr6vrrr9dXX33lc8w//vGPkqSuXbvK5XKVWNvxtm/frsGDB6thw4Zyu91q3bq1nn/+eZ9tfvvtN917773q0KGDwsLCVLduXcXFxendd98tdn+FhYV67rnnvM+52rVrq1u3blq8eHGxbZcvX66LLrpIHo9H559/frE3fX9ER0erQYMG2rt3b5nvIyAgQDfffLPmzp2rwsJC7/jLL7+sqKgoXXnllaW+r3/+85+Ki4tTUFCQGjdurIceekgvvfRSsY/tij6CWrRokTp27KigoCBNmDBBkvTFF1+od+/eqlOnjoKCgtShQwfv61KRk30U+NFHHxX7iCgxMVFt27bVJ598om7dusnj8XhrKygoKP0DhZMzVEq//vqrhYWFWZcuXczM7KWXXjJJNmfOHO8299xzj3k8HsvOzvbZ94UXXjBJ9vnnn5uZ2aFDh6xDhw5Wv359mzx5sq1atcqeeeYZCwsLs8svv9wKCwu9+0qyxo0bW/v27S01NdVWr15tX3zxhZmZDR061GbNmmVpaWmWlpZmjz32mHk8HpswYYLP8QcNGmQBAQH217/+1VauXGlTpkyxqKgoCwsLsyFDhni327Nnj0VFRVl0dLRNnz7dVq1aZY899pi53W4bOnToaR+jhIQEu+CCC4qNd+7c2WrXrm1Hjx71jt16661WvXp1u/fee2358uWWmppq559/voWHh1tWVpZ3u5deeslcLpclJiZaamqqrVq1yl544QUbMWKEd5vVq1dbjRo17JJLLrGFCxfa8uXLbejQoSbJZs+e7d1u9uzZJsmaN29ud911l61YscJeeuklq1Onjl122WU+NT/00EMmyfr27WtvvPGGrVy50iZPnmwPPfSQmZnt3bvX3G63jRs3zme/n376yTwej91///0nfZz2799vNWrUKLbv0aNHLTIy0vr27esdmzZtmqWkpNjixYttzZo1NnfuXLvwwgutVatWlp+fX6y3nTt3eseio6N9fr5FEhISLCEhwXvbn+djSZKTk02SDRo0yN5//32bN2+eNW/e3MLCwuybb74xM7OMjAz729/+5v2ZrFu3zr799tuT3mdGRoaFhYVZu3btbN68ebZy5Uq79957LSAgwMaPH+/d7uDBgzZ06FCbP3++rV692pYvX2733XefBQQE2Ny5c33u86abbjKXy2V//vOf7d1337Vly5bZ3//+d3vmmWd8HrMmTZpYmzZtbN68ebZixQrr37+/SbI1a9ac8nEwO/b7escdd/iMHTx40KpVq2a9evXyGS/6ma1fv96OHDnisxwvOjrarrnmGvv222/N5XLZ0qVLzezY86Vx48b28MMP2xtvvGGS7MMPPzxlfVu2bLGgoCBr3769vfbaa7Z48WK7+uqrrVmzZiU+fxo1amTNmze3l19+2T788EPbsGGDff311xYSEmItWrSwefPm2fvvv2+DBg0ySfbEE08U6+/4+zQz+/DDD4vVmpCQYPXq1bPIyEh79tlnbcWKFTZy5MgSH0+UDeGjkpo3b55JshdffNHMzHJzc61WrVp2ySWXeLf5/PPPTZLNmDHDZ98//OEP1qlTJ+/tlJQUCwgIsPT0dJ/t3nzzTZPkfXExO/ZiFhYWZj///PMp6ysoKLAjR47Yo48+avXq1fO+YWRkZJgke+CBB3y2f/XVV02Sz5vT8OHDrVatWvb999/7bPuPf/zDJFlGRsYpaygKH0UvoHv27LGHH37Y53EzM1u3bp1Jsqeeespn/8zMTPN4PDZmzBgzO/YYh4aG2sUXX3zKN8Dzzz/fOnbsWOxF+9prr7VGjRpZQUGBmf3vxfD44GJmNmnSJJNke/bsMTOzHTt2WLVq1ezGG288Zb9Dhgyxhg0bWl5ennfsiSeesICAgGIvuCfq27evNWnSxFubmdnSpUtNki1ZsqTEfQoLC+3IkSP2/fffmyR79913vevOJHz483w80YEDB8zj8djVV1/tM75r1y5zu902ePDgYjWeeJySXHXVVdakSZNiQf7OO++0oKCgk/4+HD161I4cOWJ/+tOfrGPHjt7xjz/+2CQVC3wnio6OtqCgIJ/fgcOHD1vdunVt+PDhp6276Pl15MgRy8/Pt2+++cauu+46CwkJsY0bN/psW/R4lLQc/1wuCh9mx352/fr1MzOz999/31wul+3cubPU4aN///5Ws2ZN279/v3esoKDA2rRpU+Lzp1q1arZt2zaf+xg4cKC53W7btWuXz3jPnj0tODjYDh486NNfacPHic9ps2P/SQkICCj2mgT/ET4qqYSEBPN4PN5fLDOzW265xSR5/3dnZtapUyeLi4vz3v7yyy9Nkj3//PPese7du1v79u2L/W8nNzfXXC6X983X7NiL2fXXX19iTR988IFdccUVFhoaWuzFq2j2oGjWZdOmTT77HjlyxAIDA33enBo3bmy9evUqVldRgHnhhRdO+xiV9EI6duxYn+3GjRtnLpfL9u7dW+xY3bp1sz/84Q9mZrZixQqTZKmpqSc95vbt202S/eMf/yh2X0W9f/nll2b2vxfD5cuX+9zH8uXLvf8DNTObPn26SbK1a9eest9///vfJskWLFhgZsdexJs1a1bsf7glWbJkiUmyFStWeMf69+9vERERPjNEe/futeHDh1uTJk0sICDA53GdOHGid7szCR/+PB9PVBSYXn/99WLrevbsaeHh4cVqPF34OHz4sAUGBtpdd91VrKai4x0fiF5//XWLj4+3mjVr+jw+QUFB3m3Gjh1rkmz37t2nPHZ0dLR169at2Hi3bt3s//2//3fKfc2sxOd/9erV7b333iu2bdHjMW/ePEtPT/dZTqypKHzMnTvXatSoYT/++KP17dvXLr/8cjOzUoePhg0blvj8HD9+fInPn+MD3PH3cWLYNDNbuHChSbJly5b59Ffa8BESElLsPou2nT9//in7wulxzkcl9O233+rjjz/WNddcIzPTwYMHdfDgQfXr10+SfD4PHjZsmNatW6evv/5a0rEz2d1utwYNGuTdZu/evfr8889VvXp1nyUkJERmVuzz/EaNGhWracOGDUpKSpIkzZw5U59++qnS09M1btw4Sf87Meynn36SdOwkuOMFBgaqXr16PmN79+7VkiVLitV1wQUXSFKpzjNo0aKF0tPTtWHDBr3xxhu68MILlZKSotdee83nOGam8PDwYsdav3699zj79++XJDVp0uSkxyv6HP2+++4rdl8jRowose4T+3a73ZL+95iV5riS1LFjR11yySXe8xDee+89fffdd7rzzjtPuZ8k9ezZU40aNdLs2bMlSQcOHNDixYt18803e0/CLCwsVFJSkhYtWqQxY8bogw8+0IYNG7R+/Xqfes+Uv8/H4xU9v0p6jkZGRnrX++Onn37S0aNH9dxzzxWr6eqrr5b0v5/pokWLNGDAADVu3FgLFizQunXrlJ6ermHDhvmcc7N//35Vq1ZNERERpz3+ic8P6dhzpLSP94ABA5Senq61a9dq+vTpCgkJ0cCBA7V9+/YSt2/durU6d+7ss5xMv379FBQUpKefflpLlizRn/70p1LVVOSnn34q9logFX99KFLSz/Wnn3466c+7aH1ZlFRD0c+rrPeJ/+FS20ro5ZdflpnpzTff1Jtvvlls/dy5c/X444+rWrVqGjRokEaPHq05c+bo73//u+bPn68+ffqoTp063u3r168vj8dz0pPY6tev73Pb5XIV2+a1115T9erV9d577ykoKMg7/s477/hsV/RCunfvXjVu3Ng7fvTo0WK/0PXr11f79u3197//vcS6Trz8ryRBQUHeF88uXbrosssu0wUXXKBRo0bp2muvVa1atVS/fn25XC598skn3jf+4xWNNWjQQJL0ww8/nPR4RY/V2LFj1bdv3xK3adWq1WnrPt7xx42KijrltiNHjlT//v3173//W1OnTtV5552nHj16nPYY1apV00033aRnn31WBw8eVGpqqvLy8nTLLbd4t/niiy+0ZcsWzZkzR0OGDPGOl3SiakmCgoKUl5dXbPzHH3/0eY75+3w8XtHza8+ePcXW7d69+5T7nkydOnW8j88dd9xR4jYxMTGSpAULFigmJkYLFy70+T05se8GDRqooKBAWVlZJb5xlqcGDRp4fwfi4uLUunVrJSQk6J577tF77713RvcdHBysgQMHKiUlRaGhoSd9zp9MvXr1SjzxNSsrq8TtS3rtqVev3kl/3tL/ni9Fr0sn/ixOFmZPVVdJgRD+IXxUMgUFBZo7d65atGihl156qdj69957T0899ZSWLVuma6+9VnXq1FGfPn00b948xcXFKSsrS8OGDfPZ59prr1VycrLq1avnfRH1l8vlUmBgoM+liocPH9b8+fN9trv00kslSQsXLtRFF13kHX/zzTeLXcFy7bXXaunSpWrRooVPWDoT9erV08SJE3XLLbfoueee09ixY3Xttddq4sSJ+u9//6sBAwacdN/4+HiFhYXpxRdf1MCBA0t8IWzVqpViY2O1ZcsWJScnl0vNSUlJqlatmqZNm6a4uLhTbnv99deradOmuvfee7VmzRo9/fTTJdZZkltuuUWTJk3Sq6++qjlz5iguLs7n+x6K7ufEgHaySzZP1KxZM33++ec+Y9988422bdvmEwrO5PkYFxcnj8ejBQsWqH///t7xH374QatXr/bODvojODhYl112mT777DO1b99eNWrUOOm2LpfL+4VlRbKysopd7dKzZ0+lpKRo2rRpevTRR/2u6Uxccskl3itV1q1bd9rn1Oncfvvt2rt3rxISEnz+41EaCQkJWrp0qU8ALSws1BtvvFHq+7jiiiv09ttva/fu3T7/IZk3b56Cg4PVrVs3Sceef5L0+eef+/wHoKSriyQpNzdXixcv1nXXXecdS01NVUBAgPd1DGegQj/0gd+KPps//izu4+3fv9/cbrf16dPHO1Z0rkKTJk2KnVRoZvbLL79Yx44drUmTJvbUU09ZWlqarVixwmbOnGn9+/f3nntgVvLZ82bHzveQZP369bOVK1faq6++ap06dbLY2Nhin7MOGjTIqlWrZmPHjrW0tDSfq11uueUW73a7d++26OhoO//88+2FF16wDz74wN5//317/vnn7ZprrrHMzMxTPlYnu9qloKDA2rVrZ3Xr1vWeQPiXv/zFgoOD7f7777clS5bY6tWr7ZVXXrHbb7/d59ySoquKLr/8cnv11Vdt9erVNmPGDJ/HZPXq1eZ2uy0pKclSU1NtzZo19vbbb1tycrL35Dyzk59zUNJn0EVXu/Tr18/eeustW7VqlT377LP28MMPF+vviSeeMElWs2ZNn3OCSiMuLs6ioqJKPFE5Pz/fWrRoYdHR0ZaammrLly+3O+64w8477zyTZI888kix3o7/uS9YsMAk2e23326rVq2yWbNmWatWraxRo0Y+53z483wsSdHVLjfddJMtXbrU5s+fby1btvS52uX4GktzwmlGRobVqVPH/vCHP9js2bPtww8/tMWLF9vkyZN9rkx6+eWXvT1+8MEHNmfOHGvRooX39+B4RVe7/OUvf7HFixfbihUrbOLEifbss896tzn+/IrjnXiezMmc7Pd1165dFhQUZFdccYXfj8fJajpeac/52Lx5s/dql4ULF3qvdomOjjZJPid2nuy4RVe7nHfeebZgwQJbunSp3XjjjSbJJk2a5N3u6NGj1qpVK2vatKmlpqbasmXL7C9/+YvFxMSc8mqX5557zlasWGF3332392eLM0f4qGT69OljNWrUsH379p10m4EDB1pgYKD3JM+CggLvG8rJzq7/5Zdf7G9/+5u1atXKatSo4b2s8J577vG51PRkL2Zmx154W7VqZW6325o3b24pKSk2a9asYm9Cv/32m40ePdoaNmxoQUFB1q1bN1u3bp2FhYXZPffc43Of+/fvt5EjR1pMTIxVr17d6tata506dbJx48bZL7/8csrH6mThw+zYmfmSfC4Dfvnll61r165Ws2ZN83g81qJFC7v55puLXRWwdOlSS0hIsJo1a1pwcLC1adOmWBjcsmWLDRgwwBo2bGjVq1e3iIgIu/zyy32usvEnfJgdu8KpS5cuFhQUZLVq1bKOHTv6XLpb5LvvvjNJdtttt53y8SnJjBkzTFKJl2ibHTthuUePHhYSEmJ16tSx/v37265du0oVPgoLC23SpEnWvHlzCwoKss6dO9vq1atLfCMt7fPxZF566SVr3769d9/evXsXuzrKn/BhZrZz504bNmyYNW7c2KpXr24NGjSw+Ph4e/zxx322mzhxojVr1szcbre1bt3aZs6caY888kix8FFQUGBPP/20tW3b1ltnXFycz9VFZyt8mJndf//9PpfsVkT4MDP75JNPrGvXruZ2uy0iIsLuv/9+b4A+Pjyf6rhbt261Xr16WVhYmNWoUcMuvPDCEn83vvnmG0tKSrLQ0FBr0KCB3XXXXd7XghPDxwUXXGAfffSRde7c2dxutzVq1MgefPDBYlexoWxcZid8AxRQAdauXavu3bvrlVdeKfErn1F6zz33nEaOHKkvvvjCe3IuUJkkJSXpu+++0zfffFMhx09MTNSPP/6oL774okKO/3vAOR9wXFpamtatW6dOnTrJ4/Foy5YtmjhxomJjY/0+YQ3/89lnn2nnzp169NFH1bt3b4IHKoXRo0erY8eOioqK0s8//6xXXnlFaWlpFfLnIuAcwgccFxoaqpUrV2rKlCnKzc1V/fr1vSfg+XvCGv7n+uuvV1ZWli655BK9+OKLFV0OUCoFBQV6+OGHlZWVJZfLpTZt2mj+/Pner79H1cTHLgAAwFF8yRgAAHAU4QMAADiK8AEAABx1zp1wWlhYqN27dyskJKTU38wIAAAqlpkpNzdXkZGRCgg49dzGORc+du/efdq/XwEAAM5NmZmZp/1DmOdc+AgJCZF0rPjQ0NAKrgYAAJRGTk6OoqKivO/jp3LOhY+ij1pCQ0MJHwAAVDKlOWWCE04BAICjCB8AAMBRhA8AAOAowgcAAHAU4QMAADiK8AEAABxF+AAAAI4ifAAAAEcRPgAAgKMIHwAAwFGEDwAA4CjCBwAAcBThAwAAOIrwAQAAHEX4AAAAjgqs6AIAAEAZuFxn/xhmZ+VumfkAAACOInwAAABHET4AAICjCB8AAMBRhA8AAOAowgcAAHAU4QMAADiK8AEAABxF+AAAAI4ifAAAAEcRPgAAgKMIHwAAwFGEDwAA4CjCBwAAcBThAwAAOMqv8NGsWTO5XK5iyx133CFJMjONHz9ekZGR8ng8SkxMVEZGxlkpHAAAVE5+hY/09HTt2bPHu6SlpUmS+vfvL0maNGmSJk+erKlTpyo9PV0RERHq0aOHcnNzy79yAABQKfkVPho0aKCIiAjv8t5776lFixZKSEiQmWnKlCkaN26c+vbtq7Zt22ru3Ln69ddflZqaerbqBwAAlUyZz/nIz8/XggULNGzYMLlcLu3cuVNZWVlKSkrybuN2u5WQkKC1a9ee9H7y8vKUk5PjswAAgKqrzOHjnXfe0cGDBzV06FBJUlZWliQpPDzcZ7vw8HDvupKkpKQoLCzMu0RFRZW1JAAAUAmUOXzMmjVLPXv2VGRkpM+4y+XyuW1mxcaON3bsWGVnZ3uXzMzMspYEAAAqgcCy7PT9999r1apVWrRokXcsIiJC0rEZkEaNGnnH9+3bV2w25Hhut1tut7ssZQAAgEqoTDMfs2fPVsOGDXXNNdd4x2JiYhQREeG9AkY6dl7ImjVrFB8ff+aVAgCAKsHvmY/CwkLNnj1bQ4YMUWDg/3Z3uVwaNWqUkpOTFRsbq9jYWCUnJys4OFiDBw8u16IBAEDl5Xf4WLVqlXbt2qVhw4YVWzdmzBgdPnxYI0aM0IEDB9S1a1etXLlSISEh5VIsAACo/FxmZhVdxPFycnIUFham7OxshYaGVnQ5AACcm05xMUe58SMi+PP+zd92AQAAjiJ8AAAARxE+AACAowgfAADAUYQPAADgKMIHAABwFOEDAAA4qkx/2wUAgErtbH9Hxrn1FVrnHMIHAKD0zrEvtkLlxMcuAADAUYQPAADgKMIHAABwFOEDAAA4ivABAAAcRfgAAACOInwAAABHET4AAICjCB8AAMBRhA8AAOAowgcAAHAU4QMAADiK8AEAABzFX7UFcO7jL6kCVQozHwAAwFGEDwAA4CjCBwAAcBThAwAAOIoTToGqjBM1AZyDmPkAAACOYuYDOJmzPWvAjMHvC7NQgBczHwAAwFGEDwAA4Cg+dkH5Y3oZAHAKzHwAAABHET4AAICjCB8AAMBRfoeP//73v/rjH/+oevXqKTg4WB06dNCmTZu8681M48ePV2RkpDwejxITE5WRkVGuRVdZLtfZXwAAqGB+hY8DBw6oe/fuql69upYtW6Yvv/xSTz31lGrXru3dZtKkSZo8ebKmTp2q9PR0RUREqEePHsrNzS3v2gEAQCXkMiv9ZQN//etf9emnn+qTTz4pcb2ZKTIyUqNGjdIDDzwgScrLy1N4eLieeOIJDR8+/LTHyMnJUVhYmLKzsxUaGlra0qqGqnKVCH2UTlXoQaKP0qoKPUj0UVpVoQfJrz78ef/2a+Zj8eLF6ty5s/r376+GDRuqY8eOmjlzpnf9zp07lZWVpaSkJO+Y2+1WQkKC1q5d68+h/MdHFgAAVAp+hY8dO3Zo2rRpio2N1YoVK3Tbbbdp5MiRmjdvniQpKytLkhQeHu6zX3h4uHfdifLy8pSTk+OzAACAqsuvLxkrLCxU586dlZycLEnq2LGjMjIyNG3aNN18883e7VwnzBKYWbGxIikpKZowYYK/dQMAgErKr5mPRo0aqU2bNj5jrVu31q5duyRJERERklRslmPfvn3FZkOKjB07VtnZ2d4lMzPTn5IAAEAl41f46N69u7Zt2+Yz9s033yg6OlqSFBMTo4iICKWlpXnX5+fna82aNYqPjy/xPt1ut0JDQ30WAABQdfn1scs999yj+Ph4JScna8CAAdqwYYNmzJihGTNmSDr2ccuoUaOUnJys2NhYxcbGKjk5WcHBwRo8ePBZaQAAAFQufoWPLl266O2339bYsWP16KOPKiYmRlOmTNGNN97o3WbMmDE6fPiwRowYoQMHDqhr165auXKlQkJCyr14AABQ+fj1PR9OKPP3fJxj1zuXSVXoQaKP0qoKPUj0UVpVoQeJPkqrKvQgnRvf8wEAAHCmCB8AAMBRhA8AAOAowgcAAHAU4QMAADiK8AEAABxF+AAAAI4ifAAAAEcRPgAAgKMIHwAAwFGEDwAA4CjCBwAAcBThAwAAOIrwAQAAHEX4AAAAjiJ8AAAARxE+AACAowgfAADAUYQPAADgKMIHAABwFOEDAAA4ivABAAAcRfgAAACOInwAAABHET4AAICjCB8AAMBRhA8AAOAowgcAAHAU4QMAADiK8AEAABxF+AAAAI4ifAAAAEcRPgAAgKMIHwAAwFGEDwAA4CjCBwAAcBThAwAAOMqv8DF+/Hi5XC6fJSIiwrvezDR+/HhFRkbK4/EoMTFRGRkZ5V40AACovPye+bjgggu0Z88e77J161bvukmTJmny5MmaOnWq0tPTFRERoR49eig3N7dciwYAAJWX3+EjMDBQERER3qVBgwaSjs16TJkyRePGjVPfvn3Vtm1bzZ07V7/++qtSU1PLvXAAAFA5+R0+tm/frsjISMXExGjgwIHasWOHJGnnzp3KyspSUlKSd1u3262EhAStXbv2pPeXl5ennJwcnwUAAFRdfoWPrl27at68eVqxYoVmzpyprKwsxcfH66efflJWVpYkKTw83Gef8PBw77qSpKSkKCwszLtERUWVoQ0AAFBZ+BU+evbsqRtuuEHt2rXTlVdeqffff1+SNHfuXO82LpfLZx8zKzZ2vLFjxyo7O9u7ZGZm+lMSAACoZM7oUtuaNWuqXbt22r59u/eqlxNnOfbt21dsNuR4brdboaGhPgsAAKi6zih85OXl6auvvlKjRo0UExOjiIgIpaWledfn5+drzZo1io+PP+NCAQBA1RDoz8b33XefevXqpaZNm2rfvn16/PHHlZOToyFDhsjlcmnUqFFKTk5WbGysYmNjlZycrODgYA0ePPhs1Q8AACoZv8LHDz/8oEGDBunHH39UgwYN1K1bN61fv17R0dGSpDFjxujw4cMaMWKEDhw4oK5du2rlypUKCQk5K8UDAIDKx2VmVtFFHC8nJ0dhYWHKzs727/yPU5zUWm7O9kNVFXqQ6KO0qkIPEn2UVlXoQaKP0qoKPUh+9eHP+zd/2wUAADiK8AEAABxF+AAAAI4ifAAAAEcRPgAAgKMIHwAAwFGEDwAA4CjCBwAAcBThAwAAOIrwAQAAHEX4AAAAjiJ8AAAARxE+AACAowgfAADAUYQPAADgKMIHAABwFOEDAAA4ivABAAAcRfgAAACOInwAAABHET4AAICjCB8AAMBRhA8AAOAowgcAAHAU4QMAADiK8AEAABxF+AAAAI4ifAAAAEcRPgAAgKMIHwAAwFGEDwAA4CjCBwAAcBThAwAAOIrwAQAAHEX4AAAAjiJ8AAAAR51R+EhJSZHL5dKoUaO8Y2am8ePHKzIyUh6PR4mJicrIyDjTOgEAQBVR5vCRnp6uGTNmqH379j7jkyZN0uTJkzV16lSlp6crIiJCPXr0UG5u7hkXCwAAKr8yhY9ffvlFN954o2bOnKk6dep4x81MU6ZM0bhx49S3b1+1bdtWc+fO1a+//qrU1NRyKxoAAFReZQofd9xxh6655hpdeeWVPuM7d+5UVlaWkpKSvGNut1sJCQlau3ZtifeVl5ennJwcnwUAAFRdgf7u8Nprr+nf//630tPTi63LysqSJIWHh/uMh4eH6/vvvy/x/lJSUjRhwgR/ywAAAJWUXzMfmZmZuvvuu7VgwQIFBQWddDuXy+Vz28yKjRUZO3assrOzvUtmZqY/JQEAgErGr5mPTZs2ad++ferUqZN3rKCgQB9//LGmTp2qbdu2STo2A9KoUSPvNvv27Ss2G1LE7XbL7XaXpXYAAFAJ+TXzccUVV2jr1q3avHmzd+ncubNuvPFGbd68Wc2bN1dERITS0tK8++Tn52vNmjWKj48v9+IBAEDl49fMR0hIiNq2beszVrNmTdWrV887PmrUKCUnJys2NlaxsbFKTk5WcHCwBg8eXH5VAwCASsvvE05PZ8yYMTp8+LBGjBihAwcOqGvXrlq5cqVCQkLK+1AAAKAScpmZVXQRx8vJyVFYWJiys7MVGhpa+h1PckJruTrbD1VV6EGij9KqCj1I9FFaVaEHiT5Kqyr0IPnVhz/v3/xtFwAA4CjCBwAAcBThAwAAOIrwAQAAHEX4AAAAjiJ8AAAARxE+AACAowgfAADAUYQPAADgKMIHAABwFOEDAAA4ivABAAAcRfgAAACOInwAAABHET4AAICjCB8AAMBRhA8AAOAowgcAAHAU4QMAADiK8AEAABxF+AAAAI4ifAAAAEcRPgAAgKMIHwAAwFGEDwAA4CjCBwAAcBThAwAAOIrwAQAAHEX4AAAAjiJ8AAAARxE+AACAowgfAADAUYQPAADgKMIHAABwFOEDAAA4ivABAAAc5Vf4mDZtmtq3b6/Q0FCFhoYqLi5Oy5Yt8643M40fP16RkZHyeDxKTExURkZGuRcNAAAqL7/CR5MmTTRx4kRt3LhRGzdu1OWXX67evXt7A8akSZM0efJkTZ06Venp6YqIiFCPHj2Um5t7VooHAACVj8vM7EzuoG7dunryySc1bNgwRUZGatSoUXrggQckSXl5eQoPD9cTTzyh4cOHl+r+cnJyFBYWpuzsbIWGhpa+EJerLOX758weqtOrCj1I9FFaVaEHiT5Kqyr0INFHaVWFHiS/+vDn/bvM53wUFBTotdde06FDhxQXF6edO3cqKytLSUlJ3m3cbrcSEhK0du3ash4GAABUMYH+7rB161bFxcXpt99+U61atfT222+rTZs23oARHh7us314eLi+//77k95fXl6e8vLyvLdzcnL8LQkAAFQifs98tGrVSps3b9b69et1++23a8iQIfryyy+9610nTAOZWbGx46WkpCgsLMy7REVF+VsSAACoRPwOHzVq1FDLli3VuXNnpaSk6MILL9QzzzyjiIgISVJWVpbP9vv27Ss2G3K8sWPHKjs727tkZmb6WxIAAKhEzvh7PsxMeXl5iomJUUREhNLS0rzr8vPztWbNGsXHx590f7fb7b10t2gBAABVl1/nfDz44IPq2bOnoqKilJubq9dee00fffSRli9fLpfLpVGjRik5OVmxsbGKjY1VcnKygoODNXjw4LNVPwAAqGT8Ch979+7VTTfdpD179igsLEzt27fX8uXL1aNHD0nSmDFjdPjwYY0YMUIHDhxQ165dtXLlSoWEhJyV4gEAQOVzxt/zUd74no+z7Hd47XmZ8T0ApUMfpVMVepDoo7SqQg/Sufc9HwAAAGVB+AAAAI4ifAAAAEcRPgAAgKMIHwAAwFGEDwAA4CjCBwAAcBThAwAAOIrwAQAAHEX4AAAAjiJ8AAAARxE+AACAowgfAADAUYQPAADgKMIHAABwFOEDAAA4ivABAAAcRfgAAACOInwAAABHET4AAICjCB8AAMBRhA8AAOAowgcAAHAU4QMAADiK8AEAABxF+AAAAI4ifAAAAEcRPgAAgKMIHwAAwFGEDwAA4CjCBwAAcBThAwAAOIrwAQAAHEX4AAAAjiJ8AAAARxE+AACAowgfAADAUX6Fj5SUFHXp0kUhISFq2LCh+vTpo23btvlsY2YaP368IiMj5fF4lJiYqIyMjHItGgAAVF5+hY81a9bojjvu0Pr165WWlqajR48qKSlJhw4d8m4zadIkTZ48WVOnTlV6eroiIiLUo0cP5ebmlnvxAACg8nGZmZV15/3796thw4Zas2aNLr30UpmZIiMjNWrUKD3wwAOSpLy8PIWHh+uJJ57Q8OHDT3ufOTk5CgsLU3Z2tkJDQ/3oxFXWNkqv7A9V6VSFHiT6KK2q0INEH6VVFXqQ6KO0qkIPkl99+PP+fUbnfGRnZ0uS6tatK0nauXOnsrKylJSU5N3G7XYrISFBa9euLfE+8vLylJOT47MAAICqq8zhw8w0evRoXXzxxWrbtq0kKSsrS5IUHh7us214eLh33YlSUlIUFhbmXaKiospaEgAAqATKHD7uvPNOff7553r11VeLrXOdMBVkZsXGiowdO1bZ2dneJTMzs6wlAQCASiCwLDvdddddWrx4sT7++GM1adLEOx4RESHp2AxIo0aNvOP79u0rNhtSxO12y+12l6UMAABQCfk182FmuvPOO7Vo0SKtXr1aMTExPutjYmIUERGhtLQ071h+fr7WrFmj+Pj48qkYAABUan7NfNxxxx1KTU3Vu+++q5CQEO95HGFhYfJ4PHK5XBo1apSSk5MVGxur2NhYJScnKzg4WIMHDz4rDQAAgMrFr/Axbdo0SVJiYqLP+OzZszV06FBJ0pgxY3T48GGNGDFCBw4cUNeuXbVy5UqFhISUS8EAAKByO6Pv+Tgb+J6Ps+x3eO15mfE9AKVDH6VTFXqQ6KO0qkIP0rn5PR8AAAD+InwAAABHET4AAICjCB8AAMBRhA8AAOAowgcAAHAU4QMAADiK8AEAABxF+AAAAI4ifAAAAEcRPgAAgKMIHwAAwFGEDwAA4CjCBwAAcBThAwAAOIrwAQAAHEX4AAAAjiJ8AAAARxE+AACAowgfAADAUYQPAADgKMIHAABwFOEDAAA4ivABAAAcRfgAAACOInwAAABHET4AAICjCB8AAMBRhA8AAOAowgcAAHAU4QMAADiK8AEAABxF+AAAAI4ifAAAAEcRPgAAgKMIHwAAwFF+h4+PP/5YvXr1UmRkpFwul9555x2f9Wam8ePHKzIyUh6PR4mJicrIyCivegEAQCXnd/g4dOiQLrzwQk2dOrXE9ZMmTdLkyZM1depUpaenKyIiQj169FBubu4ZFwsAACq/QH936Nmzp3r27FniOjPTlClTNG7cOPXt21eSNHfuXIWHhys1NVXDhw8/s2oBAEClV67nfOzcuVNZWVlKSkryjrndbiUkJGjt2rUl7pOXl6ecnByfBQAAVF3lGj6ysrIkSeHh4T7j4eHh3nUnSklJUVhYmHeJiooqz5IAAMA55qxc7eJyuXxum1mxsSJjx45Vdna2d8nMzDwbJQEAgHOE3+d8nEpERISkYzMgjRo18o7v27ev2GxIEbfbLbfbXZ5lAACAc1i5znzExMQoIiJCaWlp3rH8/HytWbNG8fHx5XkoAABQSfk98/HLL7/o22+/9d7euXOnNm/erLp166pp06YaNWqUkpOTFRsbq9jYWCUnJys4OFiDBw8u18IBAEDl5Hf42Lhxoy677DLv7dGjR0uShgwZojlz5mjMmDE6fPiwRowYoQMHDqhr165auXKlQkJCyq9qAABQabnMzCq6iOPl5OQoLCxM2dnZCg0NLf2OJzmhtVyd7YeqKvQg0UdpVYUeJPoorarQg0QfpVUVepD86sOf92/+tgsAAHAU4QMAADiK8AEAABxF+AAAAI4ifAAAAEcRPgAAgKMIHwAAwFGEDwAA4CjCBwAAcBThAwAAOIrwAQAAHEX4AAAAjiJ8AAAARxE+AACAowgfAADAUYQPAADgKMIHAABwFOEDAAA4ivABAAAcRfgAAACOInwAAABHET4AAICjCB8AAMBRhA8AAOAowgcAAHAU4QMAADiK8AEAABxF+AAAAI4ifAAAAEcRPgAAgKMIHwAAwFGEDwAA4CjCBwAAcBThAwAAOIrwAQAAHEX4AAAAjjpr4eOFF15QTEyMgoKC1KlTJ33yySdn61AAAKASOSvhY+HChRo1apTGjRunzz77TJdccol69uypXbt2nY3DAQCASsRlZlbed9q1a1dddNFFmjZtmnesdevW6tOnj1JSUk65b05OjsLCwpSdna3Q0NDSH9TlKmu5pVf+D5WvqtCDRB+lVRV6kOijtKpCDxJ9lFZV6EHyqw9/3r/LfeYjPz9fmzZtUlJSks94UlKS1q5dW96HAwAAlUxged/hjz/+qIKCAoWHh/uMh4eHKysrq9j2eXl5ysvL897Ozs6WdCxBnXPOxZr8VRV6kKpGH1WhB4k+ziVVoQepavRRFXqQ/Oqj6H27NB+olHv4KOI6YTrIzIqNSVJKSoomTJhQbDwqKupslVZ2YWEVXcGZqwo9SFWjj6rQg0Qf55Kq0INUNfqoCj1IZeojNzdXYafZr9zDR/369VWtWrVisxz79u0rNhsiSWPHjtXo0aO9twsLC/Xzzz+rXr16JYaV8pCTk6OoqChlZmb6d17JOaYq9FEVepDo41xSFXqQqkYfVaEHiT5Ky8yUm5uryMjI025b7uGjRo0a6tSpk9LS0nT99dd7x9PS0tS7d+9i27vdbrndbp+x2rVrl3dZJQoNDa3UT6QiVaGPqtCDRB/nkqrQg1Q1+qgKPUj0URqnm/EoclY+dhk9erRuuukmde7cWXFxcZoxY4Z27dql22677WwcDgAAVCJnJXz83//9n3766Sc9+uij2rNnj9q2baulS5cqOjr6bBwOAABUImfthNMRI0ZoxIgRZ+vuz4jb7dYjjzxS7OOeyqYq9FEVepDo41xSFXqQqkYfVaEHiT7OhrPyJWMAAAAnwx+WAwAAjiJ8AAAARxE+AACAowgfAADAUYSPSo7zhQEAlc1Zu9QWznC73dqyZYtat25d0aUAOAN79uzRtGnT9M9//lN79uxRtWrVFBMToz59+mjo0KGqVq1aRZcIlJvfxaW2X331ldavX6+4uDidf/75+vrrr/XMM88oLy9Pf/zjH3X55ZdXdImndfzfvzneM888oz/+8Y+qV6+eJGny5MlOlnXGDhw4oLlz52r79u1q1KiRhgwZcm7+UcHjfPbZZ6pdu7ZiYmIkSQsWLNC0adO0a9cuRUdH684779TAgQMruMrTu+uuuzRgwABdcsklFV3KGXnuuee0ceNGXXPNNRowYIDmz5+vlJQUFRYWqm/fvnr00UcVGHhu/z9r48aNuvLKKxUTEyOPx6N//etfuvHGG5Wfn68VK1aodevWWrFihUJCQiq61NM6dOiQUlNTtXbtWmVlZcnlcik8PFzdu3fXoEGDVLNmzYou8Yzt3btX06dP18MPP1zRpZzWDz/8oNq1a6tWrVo+40eOHNG6det06aWXVkxhVsUtW7bMatSoYXXr1rWgoCBbtmyZNWjQwK688kq74oorLDAw0D744IOKLvO0XC6XdejQwRITE30Wl8tlXbp0scTERLvssssquszTatSokf34449mZrZjxw6LiIiwiIgI69GjhzVp0sTCwsLsq6++quAqT61jx462evVqMzObOXOmeTweGzlypE2bNs1GjRpltWrVslmzZlVwlafncrksICDAYmNjbeLEibZnz56KLslvjz76qIWEhNgNN9xgERERNnHiRKtXr549/vjjlpycbA0aNLCHH364oss8re7du9v48eO9t+fPn29du3Y1M7Off/7ZOnToYCNHjqyo8kotIyPDIiMjrXbt2ta7d2/7y1/+Yrfeeqv17t3bateubY0bN7aMjIyKLvOMbd682QICAiq6jFPavXu3denSxQICAqxatWp28803W25urnd9VlZWhfZQ5cNHXFycjRs3zszMXn31VatTp449+OCD3vUPPvig9ejRo6LKK7Xk5GSLiYkpFpQCAwMr1S+zy+WyvXv3mpnZwIEDLTEx0Q4dOmRmZr/99ptde+211q9fv4os8bSCg4Pt+++/N7NjQWT69Ok+61955RVr06ZNRZTmF5fLZatWrbK7777b6tevb9WrV7frrrvOlixZYgUFBRVdXqk0b97c3nrrLTM79oZQrVo1W7BggXf9okWLrGXLlhVVXql5PB77z3/+471dUFBg1atXt6ysLDMzW7lypUVGRlZUeaWWmJhoAwcOtLy8vGLr8vLybNCgQZaYmFgBlflny5Ytp1wWLlx4zoePm2++2bp162bp6emWlpZmnTt3tk6dOtnPP/9sZsfCh8vlqrD6qnz4CA0Nte3bt5vZsV/owMBA27Rpk3f91q1bLTw8vKLK88uGDRvsvPPOs3vvvdfy8/PNrHKHj5LC1Pr1661JkyYVUVqp1atXzzZu3GhmZg0bNrTNmzf7rP/222/N4/FURGl+Of5nkZ+fbwsXLrSrrrrKqlWrZpGRkfbggw96f3fOVR6PxxsEzcyqV69uX3zxhff2d999Z8HBwRVRml+io6Ptn//8p/f27t27zeVy2a+//mpmZjt37rSgoKCKKq/UPB7PKV+Ptm7dWml+NwICAszlchVbisbP9fARGRlp//rXv7y3f/vtN+vdu7d16NDBfvrppwqf+fhdXe0SEBCgoKAg1a5d2zsWEhKi7OzsiivKD126dNGmTZu0f/9+de7cWVu3bpXL5arosvxWVHNeXp7Cw8N91oWHh2v//v0VUVap9ezZU9OmTZMkJSQk6M033/RZ//rrr6tly5YVUVqZVa9eXQMGDNDy5cu1Y8cO3XrrrXrllVfUqlWrii7tlCIiIvTll19KkrZv366CggLvbUnKyMhQw4YNK6q8UuvTp49uu+02LV++XB9++KFuvPFGJSQkyOPxSJK2bdumxo0bV3CVp1enTh1t3779pOu//fZb1alTx8GKyqZevXqaOXOmdu7cWWzZsWOH3nvvvYou8bSys7N9Hmu3260333xTzZo102WXXaZ9+/ZVYHWq+ud8tG/f3pYtW+a9vXXrVjty5Ij39ieffGIxMTEVUdoZefXVVy08PNwCAgIq3cxHu3btrGPHjlarVi1btGiRz/o1a9ZY48aNK6i60vnvf/9rzZo1s0svvdRGjx5tHo/HLr74Yrv11lvt0ksvtRo1atj7779f0WWe1vEzHyUpLCy0lStXOliR/8aNG2cNGjSwP//5zxYTE2Njx461pk2b2rRp0+zFF1+0qKgou+eeeyq6zNPKzc21AQMGWGBgoLlcLouPj7cdO3Z4169YscJef/31CqywdB555BELCwuzJ5980jZv3mx79uyxrKws27x5sz355JNWp04dmzBhQkWXeVpXXXWVPfbYYyddv3nz5gr9yKI02rVrZ2+++Wax8SNHjlifPn2sadOmFTrzcW6fAl4Obr/9dhUUFHhvt23b1mf9smXLKsXVLicaOHCgLr74Ym3atEnR0dEVXU6pPfLIIz63g4ODfW4vWbLknL/6IjIyUp999pkmTpyoJUuWyMy0YcMGZWZmqnv37vr000/VuXPnii7ztKKjo095+abL5VKPHj0crMh/EyZMkMfj0fr16zV8+HA98MADat++vcaMGaNff/1VvXr10mOPPVbRZZ5WrVq1tHDhQv322286evRosSsTkpKSKqgy/4wfP14ej0eTJ0/WmDFjvLOcZqaIiAj99a9/1ZgxYyq4ytMbPny4Dh06dNL1TZs21ezZsx2syH89e/bUjBkzdMMNN/iMBwYG6o033tANN9ygzMzMCqrud3KpLQDAWTt37lRWVpakYx+PFV2aDmccPXpUv/76q0JDQ0tcX1BQoB9++KHC/vP6uzrnAwDgjJiYGMXFxSkuLs4bPDIzMzVs2LAKruzMVYY+AgMDTxo8JGn37t2aMGGCgxX5YuYDAOCILVu26KKLLvL5KLwyqgp9VHQPVf6cDwCAMxYvXnzK9Tt27HCokjNTFfo413tg5gMAUC4CAgLkcrlO+QcvXS7XOT9jUBX6ONd74JwPAEC5aNSokd566y0VFhaWuPz73/+u6BJLpSr0ca73QPgAAJSLTp06nfJN7XT/Ez9XVIU+zvUeOOcDAFAu7r///lN+P0bLli314YcfOlhR2VSFPs71HjjnAwAAOIqPXQAAgKMIHwAAwFGEDwAA4CjCBwAAcBThA0C5atasmaZMmVLRZQA4hxE+gCrK5XKdchk6dOhp93/nnXccqRXA7wvf8wFUUXv27PH+e+HChXr44Ye1bds275jH46mIsipMfn6+atSoUdFlABAzH0CVFRER4V3CwsLkcrl8xlJTU9WiRQvVqFFDrVq10vz58737NmvWTJJ0/fXXy+VyeW//5z//Ue/evRUeHq5atWqpS5cuWrVqVZlrHD9+vJo2bSq3263IyEiNHDnSuy4vL09jxoxRVFSU3G63YmNjNWvWLO/6NWvW6A9/+IPcbrcaNWqkv/71rzp69Kh3fWJiou68806NHj1a9evXV48ePSRJX375pa6++mrVqlVL4eHhuummm/Tjjz+WuQcA/iN8AL9Db7/9tu6++27de++9+uKLLzR8+HDdcsst3m88TE9PlyTNnj1be/bs8d7+5ZdfdPXVV2vVqlX67LPPdNVVV6lXr17atWuX3zW8+eabevrppzV9+nRt375d77zzjtq1a+ddf/PNN+u1117Ts88+q6+++kovvviiatWqJUn673//q6uvvlpdunTRli1bNG3aNM2aNUuPP/64zzHmzp2rwMBAffrpp5o+fbr27NmjhIQEdejQQRs3btTy5cu1d+9eDRgwoEyPI4AyMgBV3uzZsy0sLMx7Oz4+3m699Vafbfr3729XX32197Yke/vtt097323atLHnnnvOezs6Otqefvrp0+731FNP2XnnnWf5+fnF1m3bts0kWVpaWon7Pvjgg9aqVSsrLCz0jj3//PNWq1YtKygoMDOzhIQE69Chg89+Dz30kCUlJfmMZWZmmiTbtm3baWsGUD6Y+QB+h7766it1797dZ6x79+766quvTrnfoUOHNGbMGLVp00a1a9dWrVq19PXXX5dp5qN///46fPiwmjdvrltvvVVvv/2292OTzZs3q1q1akpISDhp/XFxcXK5XD71//LLL/rhhx+8Y507d/bZb9OmTfrwww9Vq1Yt73L++edLOvaREgBncMIp8Dt1/Bu3JJlZsbET3X///VqxYoX+8Y9/qGXLlvJ4POrXr5/y8/P9Pn5UVJS2bdumtLQ0rVq1SiNGjNCTTz6pNWvWnPZk2JJqtf//z1QdP16zZk2fbQoLC9WrVy898cQTxe6zUaNGfvcAoGyY+QB+h1q3bq1//vOfPmNr165V69atvberV6+ugoICn20++eQTDR06VNdff73atWuniIgIfffdd2Wuw+Px6LrrrtOzzz6rjz76SOvWrdPWrVvVrl07FRYWas2aNSXu16ZNG61du9bnT4KvXbtWISEhaty48UmPd9FFFykjI0PNmjVTy5YtfZYTgwqAs4fwAfwO3X///ZozZ45efPFFbd++XZMnT9aiRYt03333ebdp1qyZPvjgA2VlZenAgQOSjv0Z7kWLFmnz5s3asmWLBg8erMLCwjLVMGfOHM2aNUtffPGFduzYofnz58vj8Sg6OlrNmjXTkCFDNGzYML3zzjvauXOnPvroI73++uuSpBEjRigzM1N33XWXvv76a7377rt65JFHNHr0aAUEnPxl7Y477tDPP/+sQYMGacOGDdqxY4dWrlypYcOGFQtaAM4ewgfwO9SnTx8988wzevLJJ3XBBRdo+vTpmj17thITE73bPPXUU0pLS1NUVJQ6duwoSXr66adVp04dxcfHq1evXrrqqqt00UUXlamG2rVra+bMmerevbvat2+vDz74QEuWLFG9evUkSdOmTVO/fv00YsQInX/++br11lt16NAhSVLjxo21dOlSbdiwQRdeeKFuu+02/elPf9Lf/va3Ux4zMjJSn376qQoKCnTVVVepbdu2uvvuuxUWFnbK0AKgfLns+HlLAACAs4yoDwAAHEX4AHBWvPLKKz6XtB6/XHDBBRVdHoAKxMcuAM6K3Nxc7d27t8R11atXV3R0tMMVAThXED4AAICj+NgFAAA4ivABAAAcRfgAAACOInwAAABHET4AAICjCB8AAMBRhA8AAOAowgcAAHDU/wefI2M3QzPbTAAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "rfm_segmentation.groupby('Total_score')['Recency'].mean().plot(kind='bar', colormap='bwr_r')\n",
    "\n",
    "plt.title('Average Recency value of each RFM group')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6377426c",
   "metadata": {},
   "source": [
    "__Based on Recency we see Total_score 10,11,12 have highest value which is good for model. because it could have\n",
    "combination of values such as 444, 434, 334 etc__"
   ]
  }
 ],
 "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}