Cohort_Anaylsis_1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is the description code for customer segmentation analysis. The original resource is from the course [Customer Segmentation Analysis in Python](https://www.datacamp.com/courses/customer-segmentation-in-python) instructed by Karolis Urbonas on [DataCamp](https://www.datacamp.com). The dataset is from UCI ML repo, which is also available from [Kaggle](https://www.kaggle.com/jihyeseo/online-retail-data-set-from-uci-ml-repo)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import library\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import datetime as dt\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.cluster import KMeans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import the dataset\n",
    "online = pd.read_excel('Online Retail.xlsx')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# drop the row missing customerID\n",
    "online = online[online.CustomerID.notnull()]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I'll use only the subset of the full dataset, taking 20% of samples."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>InvoiceNo</th>\n",
       "      <th>StockCode</th>\n",
       "      <th>Description</th>\n",
       "      <th>Quantity</th>\n",
       "      <th>InvoiceDate</th>\n",
       "      <th>UnitPrice</th>\n",
       "      <th>CustomerID</th>\n",
       "      <th>Country</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>542401</td>\n",
       "      <td>22502</td>\n",
       "      <td>PICNIC BASKET WICKER SMALL</td>\n",
       "      <td>2</td>\n",
       "      <td>2011-01-27 15:51:00</td>\n",
       "      <td>5.95</td>\n",
       "      <td>14541.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>542231</td>\n",
       "      <td>22726</td>\n",
       "      <td>ALARM CLOCK BAKELIKE GREEN</td>\n",
       "      <td>2</td>\n",
       "      <td>2011-01-26 13:40:00</td>\n",
       "      <td>3.75</td>\n",
       "      <td>16714.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>556956</td>\n",
       "      <td>22090</td>\n",
       "      <td>PAPER BUNTING RETROSPOT</td>\n",
       "      <td>40</td>\n",
       "      <td>2011-06-16 09:04:00</td>\n",
       "      <td>2.55</td>\n",
       "      <td>13694.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>573874</td>\n",
       "      <td>23581</td>\n",
       "      <td>JUMBO BAG PAISLEY PARK</td>\n",
       "      <td>10</td>\n",
       "      <td>2011-11-01 12:45:00</td>\n",
       "      <td>2.08</td>\n",
       "      <td>13868.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>580742</td>\n",
       "      <td>23343</td>\n",
       "      <td>JUMBO BAG VINTAGE CHRISTMAS</td>\n",
       "      <td>200</td>\n",
       "      <td>2011-12-06 09:30:00</td>\n",
       "      <td>1.75</td>\n",
       "      <td>13694.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  InvoiceNo StockCode                   Description  Quantity  \\\n",
       "0    542401     22502    PICNIC BASKET WICKER SMALL         2   \n",
       "1    542231     22726    ALARM CLOCK BAKELIKE GREEN         2   \n",
       "2    556956     22090       PAPER BUNTING RETROSPOT        40   \n",
       "3    573874     23581        JUMBO BAG PAISLEY PARK        10   \n",
       "4    580742     23343  JUMBO BAG VINTAGE CHRISTMAS        200   \n",
       "\n",
       "          InvoiceDate  UnitPrice  CustomerID         Country  \n",
       "0 2011-01-27 15:51:00       5.95     14541.0  United Kingdom  \n",
       "1 2011-01-26 13:40:00       3.75     16714.0  United Kingdom  \n",
       "2 2011-06-16 09:04:00       2.55     13694.0  United Kingdom  \n",
       "3 2011-11-01 12:45:00       2.08     13868.0  United Kingdom  \n",
       "4 2011-12-06 09:30:00       1.75     13694.0  United Kingdom  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# use a subset of full data\n",
    "np.random.seed(306)\n",
    "online = online.sample(frac = .3).reset_index(drop = True)\n",
    "online.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0   2011-01-27\n",
       "1   2011-01-26\n",
       "2   2011-06-16\n",
       "3   2011-11-01\n",
       "4   2011-12-06\n",
       "Name: InvoiceDay, dtype: datetime64[ns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# extract year, month and day \n",
    "online['InvoiceDay'] = online.InvoiceDate.apply(lambda x: dt.datetime(x.year, x.month, x.day))\n",
    "online.InvoiceDay.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4222"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# the number of customers\n",
    "online.CustomerID.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>InvoiceNo</th>\n",
       "      <th>StockCode</th>\n",
       "      <th>Description</th>\n",
       "      <th>Quantity</th>\n",
       "      <th>InvoiceDate</th>\n",
       "      <th>UnitPrice</th>\n",
       "      <th>CustomerID</th>\n",
       "      <th>Country</th>\n",
       "      <th>InvoiceDay</th>\n",
       "      <th>CohortDay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>542401</td>\n",
       "      <td>22502</td>\n",
       "      <td>PICNIC BASKET WICKER SMALL</td>\n",
       "      <td>2</td>\n",
       "      <td>2011-01-27 15:51:00</td>\n",
       "      <td>5.95</td>\n",
       "      <td>14541.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-01-27</td>\n",
       "      <td>2011-01-24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>542231</td>\n",
       "      <td>22726</td>\n",
       "      <td>ALARM CLOCK BAKELIKE GREEN</td>\n",
       "      <td>2</td>\n",
       "      <td>2011-01-26 13:40:00</td>\n",
       "      <td>3.75</td>\n",
       "      <td>16714.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-01-26</td>\n",
       "      <td>2011-01-26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>556956</td>\n",
       "      <td>22090</td>\n",
       "      <td>PAPER BUNTING RETROSPOT</td>\n",
       "      <td>40</td>\n",
       "      <td>2011-06-16 09:04:00</td>\n",
       "      <td>2.55</td>\n",
       "      <td>13694.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-06-16</td>\n",
       "      <td>2010-12-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>573874</td>\n",
       "      <td>23581</td>\n",
       "      <td>JUMBO BAG PAISLEY PARK</td>\n",
       "      <td>10</td>\n",
       "      <td>2011-11-01 12:45:00</td>\n",
       "      <td>2.08</td>\n",
       "      <td>13868.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-11-01</td>\n",
       "      <td>2011-11-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>580742</td>\n",
       "      <td>23343</td>\n",
       "      <td>JUMBO BAG VINTAGE CHRISTMAS</td>\n",
       "      <td>200</td>\n",
       "      <td>2011-12-06 09:30:00</td>\n",
       "      <td>1.75</td>\n",
       "      <td>13694.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-12-06</td>\n",
       "      <td>2010-12-01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  InvoiceNo StockCode                   Description  Quantity  \\\n",
       "0    542401     22502    PICNIC BASKET WICKER SMALL         2   \n",
       "1    542231     22726    ALARM CLOCK BAKELIKE GREEN         2   \n",
       "2    556956     22090       PAPER BUNTING RETROSPOT        40   \n",
       "3    573874     23581        JUMBO BAG PAISLEY PARK        10   \n",
       "4    580742     23343  JUMBO BAG VINTAGE CHRISTMAS        200   \n",
       "\n",
       "          InvoiceDate  UnitPrice  CustomerID         Country InvoiceDay  \\\n",
       "0 2011-01-27 15:51:00       5.95     14541.0  United Kingdom 2011-01-27   \n",
       "1 2011-01-26 13:40:00       3.75     16714.0  United Kingdom 2011-01-26   \n",
       "2 2011-06-16 09:04:00       2.55     13694.0  United Kingdom 2011-06-16   \n",
       "3 2011-11-01 12:45:00       2.08     13868.0  United Kingdom 2011-11-01   \n",
       "4 2011-12-06 09:30:00       1.75     13694.0  United Kingdom 2011-12-06   \n",
       "\n",
       "   CohortDay  \n",
       "0 2011-01-24  \n",
       "1 2011-01-26  \n",
       "2 2010-12-01  \n",
       "3 2011-11-01  \n",
       "4 2010-12-01  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Assign a minimum InvoiceDay \n",
    "online['CohortDay'] = online.groupby('CustomerID')['InvoiceDay'].transform('min')\n",
    "online.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`InvoiceDay` is the date the customer made an order and `CohortDay` is the date that each customer made the first order. To seperate the date columns into year, month and day part by defining a funtion to extract the year, month, and day from date columns."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# define a funtion \n",
    "def get_date_int(col):\n",
    "    year = col.dt.year\n",
    "    month = col.dt.month\n",
    "    day = col.dt.day\n",
    "    return year, month, day"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "invoice_year, invoice_month, invoice_day = get_date_int(online.InvoiceDay)\n",
    "cohort_year, cohort_month, cohort_day = get_date_int(online.CohortDay)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "year_diff = invoice_year - cohort_year\n",
    "month_diff = invoice_month - cohort_month\n",
    "day_diff = invoice_day - cohort_day"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    -5\n",
       "1   -16\n",
       "2    19\n",
       "3     6\n",
       "4     0\n",
       "5    -8\n",
       "6     0\n",
       "7    12\n",
       "8     2\n",
       "9    17\n",
       "dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "month_diff.head(10)\n",
    "day_diff.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>InvoiceNo</th>\n",
       "      <th>StockCode</th>\n",
       "      <th>Description</th>\n",
       "      <th>Quantity</th>\n",
       "      <th>InvoiceDate</th>\n",
       "      <th>UnitPrice</th>\n",
       "      <th>CustomerID</th>\n",
       "      <th>Country</th>\n",
       "      <th>InvoiceDay</th>\n",
       "      <th>CohortDay</th>\n",
       "      <th>CohortIndex</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>548670</td>\n",
       "      <td>21975</td>\n",
       "      <td>PACK OF 60 DINOSAUR CAKE CASES</td>\n",
       "      <td>1</td>\n",
       "      <td>2011-04-01 15:37:00</td>\n",
       "      <td>0.55</td>\n",
       "      <td>15356.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-04-01</td>\n",
       "      <td>2010-12-06</td>\n",
       "      <td>120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>580513</td>\n",
       "      <td>23321</td>\n",
       "      <td>SMALL WHITE HEART OF WICKER</td>\n",
       "      <td>2</td>\n",
       "      <td>2011-12-04 13:59:00</td>\n",
       "      <td>1.65</td>\n",
       "      <td>14456.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-12-04</td>\n",
       "      <td>2011-07-20</td>\n",
       "      <td>134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>568531</td>\n",
       "      <td>23505</td>\n",
       "      <td>PLAYING CARDS I LOVE LONDON</td>\n",
       "      <td>3</td>\n",
       "      <td>2011-09-27 13:49:00</td>\n",
       "      <td>1.25</td>\n",
       "      <td>16713.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-09-27</td>\n",
       "      <td>2010-12-08</td>\n",
       "      <td>294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>544178</td>\n",
       "      <td>21411</td>\n",
       "      <td>GINGHAM HEART  DOORSTOP RED</td>\n",
       "      <td>3</td>\n",
       "      <td>2011-02-16 14:40:00</td>\n",
       "      <td>4.25</td>\n",
       "      <td>14543.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-02-16</td>\n",
       "      <td>2010-12-10</td>\n",
       "      <td>71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>565381</td>\n",
       "      <td>22457</td>\n",
       "      <td>NATURAL SLATE HEART CHALKBOARD</td>\n",
       "      <td>6</td>\n",
       "      <td>2011-09-02 15:23:00</td>\n",
       "      <td>2.95</td>\n",
       "      <td>16173.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-09-02</td>\n",
       "      <td>2011-09-02</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  InvoiceNo StockCode                      Description  Quantity  \\\n",
       "0    548670     21975   PACK OF 60 DINOSAUR CAKE CASES         1   \n",
       "1    580513     23321      SMALL WHITE HEART OF WICKER         2   \n",
       "2    568531     23505     PLAYING CARDS I LOVE LONDON          3   \n",
       "3    544178     21411      GINGHAM HEART  DOORSTOP RED         3   \n",
       "4    565381     22457  NATURAL SLATE HEART CHALKBOARD          6   \n",
       "\n",
       "          InvoiceDate  UnitPrice  CustomerID         Country InvoiceDay  \\\n",
       "0 2011-04-01 15:37:00       0.55     15356.0  United Kingdom 2011-04-01   \n",
       "1 2011-12-04 13:59:00       1.65     14456.0  United Kingdom 2011-12-04   \n",
       "2 2011-09-27 13:49:00       1.25     16713.0  United Kingdom 2011-09-27   \n",
       "3 2011-02-16 14:40:00       4.25     14543.0  United Kingdom 2011-02-16   \n",
       "4 2011-09-02 15:23:00       2.95     16173.0  United Kingdom 2011-09-02   \n",
       "\n",
       "   CohortDay  CohortIndex  \n",
       "0 2010-12-06          120  \n",
       "1 2011-07-20          134  \n",
       "2 2010-12-08          294  \n",
       "3 2010-12-10           71  \n",
       "4 2011-09-02            0  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# calculate the day period for each invoice date \n",
    "online['CohortIndex'] = year_diff * 365 + month_diff * 30 + day_diff \n",
    "online.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1) Cohort Matrix"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Customer retention is a very useful metric to understand how many of the all customers are still active. We can also get the average price metics and see if there are any patterns across time and across cohorts."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1-1. Retention Rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead 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>CohortDay</th>\n",
       "      <th>CohortIndex</th>\n",
       "      <th>CustomerID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>0</td>\n",
       "      <td>82</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CohortDay  CohortIndex  CustomerID\n",
       "0 2010-12-01            0          82\n",
       "1 2010-12-01            1           3\n",
       "2 2010-12-01            2           2\n",
       "3 2010-12-01            4           6\n",
       "4 2010-12-01            5           5"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cohort_data = online.groupby(['CohortDay', 'CohortIndex'])['CustomerID'].apply(pd.Series.nunique).reset_index()\n",
    "cohort_data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>CohortIndex</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>364</th>\n",
       "      <th>365</th>\n",
       "      <th>366</th>\n",
       "      <th>367</th>\n",
       "      <th>368</th>\n",
       "      <th>369</th>\n",
       "      <th>370</th>\n",
       "      <th>371</th>\n",
       "      <th>372</th>\n",
       "      <th>373</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CohortDay</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",
       "      <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",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2010-12-01</th>\n",
       "      <td>82.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>...</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-02</th>\n",
       "      <td>83.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-03</th>\n",
       "      <td>41.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-05</th>\n",
       "      <td>65.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>7.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-06</th>\n",
       "      <td>72.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-07</th>\n",
       "      <td>46.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>3.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-08</th>\n",
       "      <td>74.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-09</th>\n",
       "      <td>68.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-10</th>\n",
       "      <td>38.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-12</th>\n",
       "      <td>30.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 374 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "CohortIndex   0    1    2    3    4    5     6    7    8    9   ...   364  \\\n",
       "CohortDay                                                       ...         \n",
       "2010-12-01   82.0  3.0  2.0  NaN  6.0  5.0   2.0  6.0  7.0  4.0 ...   5.0   \n",
       "2010-12-02   83.0  NaN  NaN  NaN  1.0  3.0   3.0  6.0  4.0  NaN ...   2.0   \n",
       "2010-12-03   41.0  NaN  1.0  1.0  NaN  2.0   3.0  NaN  NaN  NaN ...   1.0   \n",
       "2010-12-05   65.0  2.0  4.0  1.0  2.0  2.0   NaN  6.0  6.0  2.0 ...   7.0   \n",
       "2010-12-06   72.0  1.0  3.0  2.0  1.0  NaN   1.0  4.0  8.0  6.0 ...   4.0   \n",
       "2010-12-07   46.0  NaN  1.0  1.0  NaN  NaN   1.0  2.0  NaN  NaN ...   3.0   \n",
       "2010-12-08   74.0  1.0  NaN  NaN  1.0  1.0  10.0  3.0  4.0  1.0 ...   3.0   \n",
       "2010-12-09   68.0  NaN  NaN  NaN  1.0  3.0   1.0  2.0  2.0  NaN ...   4.0   \n",
       "2010-12-10   38.0  NaN  NaN  1.0  3.0  1.0   1.0  3.0  NaN  NaN ...   NaN   \n",
       "2010-12-12   30.0  1.0  1.0  NaN  NaN  NaN   NaN  1.0  1.0  NaN ...   NaN   \n",
       "\n",
       "CohortIndex  365  366  367  368  369  370  371  372  373  \n",
       "CohortDay                                                 \n",
       "2010-12-01   8.0  6.0  NaN  1.0  4.0  6.0  7.0  5.0  4.0  \n",
       "2010-12-02   3.0  NaN  NaN  4.0  3.0  4.0  2.0  1.0  NaN  \n",
       "2010-12-03   NaN  2.0  2.0  3.0  1.0  2.0  NaN  NaN  NaN  \n",
       "2010-12-05   2.0  3.0  3.0  3.0  2.0  NaN  NaN  NaN  NaN  \n",
       "2010-12-06   2.0  5.0  3.0  NaN  NaN  NaN  NaN  NaN  NaN  \n",
       "2010-12-07   6.0  3.0  NaN  NaN  NaN  NaN  NaN  NaN  NaN  \n",
       "2010-12-08   4.0  2.0  NaN  NaN  NaN  NaN  NaN  NaN  NaN  \n",
       "2010-12-09   1.0  NaN  NaN  NaN  NaN  NaN  NaN  NaN  NaN  \n",
       "2010-12-10   NaN  NaN  NaN  NaN  NaN  NaN  NaN  NaN  NaN  \n",
       "2010-12-12   NaN  NaN  NaN  NaN  NaN  NaN  NaN  NaN  NaN  \n",
       "\n",
       "[10 rows x 374 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# pivot the table  \n",
    "cohort_wide = cohort_data.pivot(index = 'CohortDay', columns = 'CohortIndex', values = 'CustomerID')\n",
    "cohort_wide.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we counted the number of customers per each the first order date and the day period, the table above is showing how many orders occured for each day period at each day. For example, the first row is showing customers who's first order occured at the 1st of DEC, 2010 and each coluns are showing how many reorders of this people are. You can see that the numbers are decresaing in a exponential destribution. This could be seen with the plot with heatmap."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before plotting, change the values in a proportional form. I will divide the values by the first cohort value so that we can easily contrast the decreasing percentage for each date."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>CohortIndex</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>364</th>\n",
       "      <th>365</th>\n",
       "      <th>366</th>\n",
       "      <th>367</th>\n",
       "      <th>368</th>\n",
       "      <th>369</th>\n",
       "      <th>370</th>\n",
       "      <th>371</th>\n",
       "      <th>372</th>\n",
       "      <th>373</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CohortDay</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",
       "      <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",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2010-12-01</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.037</td>\n",
       "      <td>0.024</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.073</td>\n",
       "      <td>0.061</td>\n",
       "      <td>0.024</td>\n",
       "      <td>0.073</td>\n",
       "      <td>0.085</td>\n",
       "      <td>0.049</td>\n",
       "      <td>...</td>\n",
       "      <td>0.061</td>\n",
       "      <td>0.098</td>\n",
       "      <td>0.073</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.012</td>\n",
       "      <td>0.049</td>\n",
       "      <td>0.073</td>\n",
       "      <td>0.085</td>\n",
       "      <td>0.061</td>\n",
       "      <td>0.049</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-02</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.012</td>\n",
       "      <td>0.036</td>\n",
       "      <td>0.036</td>\n",
       "      <td>0.072</td>\n",
       "      <td>0.048</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>0.024</td>\n",
       "      <td>0.036</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.048</td>\n",
       "      <td>0.036</td>\n",
       "      <td>0.048</td>\n",
       "      <td>0.024</td>\n",
       "      <td>0.012</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-03</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.024</td>\n",
       "      <td>0.024</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.049</td>\n",
       "      <td>0.073</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>0.024</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.049</td>\n",
       "      <td>0.049</td>\n",
       "      <td>0.073</td>\n",
       "      <td>0.024</td>\n",
       "      <td>0.049</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-05</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.031</td>\n",
       "      <td>0.062</td>\n",
       "      <td>0.015</td>\n",
       "      <td>0.031</td>\n",
       "      <td>0.031</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.092</td>\n",
       "      <td>0.092</td>\n",
       "      <td>0.031</td>\n",
       "      <td>...</td>\n",
       "      <td>0.108</td>\n",
       "      <td>0.031</td>\n",
       "      <td>0.046</td>\n",
       "      <td>0.046</td>\n",
       "      <td>0.046</td>\n",
       "      <td>0.031</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-06</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.014</td>\n",
       "      <td>0.042</td>\n",
       "      <td>0.028</td>\n",
       "      <td>0.014</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.014</td>\n",
       "      <td>0.056</td>\n",
       "      <td>0.111</td>\n",
       "      <td>0.083</td>\n",
       "      <td>...</td>\n",
       "      <td>0.056</td>\n",
       "      <td>0.028</td>\n",
       "      <td>0.069</td>\n",
       "      <td>0.042</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-07</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.022</td>\n",
       "      <td>0.022</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.022</td>\n",
       "      <td>0.043</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>0.065</td>\n",
       "      <td>0.130</td>\n",
       "      <td>0.065</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-08</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.014</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.014</td>\n",
       "      <td>0.014</td>\n",
       "      <td>0.135</td>\n",
       "      <td>0.041</td>\n",
       "      <td>0.054</td>\n",
       "      <td>0.014</td>\n",
       "      <td>...</td>\n",
       "      <td>0.041</td>\n",
       "      <td>0.054</td>\n",
       "      <td>0.027</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-09</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.015</td>\n",
       "      <td>0.044</td>\n",
       "      <td>0.015</td>\n",
       "      <td>0.029</td>\n",
       "      <td>0.029</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>0.059</td>\n",
       "      <td>0.015</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-10</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.026</td>\n",
       "      <td>0.079</td>\n",
       "      <td>0.026</td>\n",
       "      <td>0.026</td>\n",
       "      <td>0.079</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-12</th>\n",
       "      <td>1.0</td>\n",
       "      <td>0.033</td>\n",
       "      <td>0.033</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.033</td>\n",
       "      <td>0.033</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 374 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "CohortIndex  0      1      2      3      4      5      6      7      8    \\\n",
       "CohortDay                                                                  \n",
       "2010-12-01   1.0  0.037  0.024    NaN  0.073  0.061  0.024  0.073  0.085   \n",
       "2010-12-02   1.0    NaN    NaN    NaN  0.012  0.036  0.036  0.072  0.048   \n",
       "2010-12-03   1.0    NaN  0.024  0.024    NaN  0.049  0.073    NaN    NaN   \n",
       "2010-12-05   1.0  0.031  0.062  0.015  0.031  0.031    NaN  0.092  0.092   \n",
       "2010-12-06   1.0  0.014  0.042  0.028  0.014    NaN  0.014  0.056  0.111   \n",
       "2010-12-07   1.0    NaN  0.022  0.022    NaN    NaN  0.022  0.043    NaN   \n",
       "2010-12-08   1.0  0.014    NaN    NaN  0.014  0.014  0.135  0.041  0.054   \n",
       "2010-12-09   1.0    NaN    NaN    NaN  0.015  0.044  0.015  0.029  0.029   \n",
       "2010-12-10   1.0    NaN    NaN  0.026  0.079  0.026  0.026  0.079    NaN   \n",
       "2010-12-12   1.0  0.033  0.033    NaN    NaN    NaN    NaN  0.033  0.033   \n",
       "\n",
       "CohortIndex    9    ...      364    365    366    367    368    369    370  \\\n",
       "CohortDay           ...                                                      \n",
       "2010-12-01   0.049  ...    0.061  0.098  0.073    NaN  0.012  0.049  0.073   \n",
       "2010-12-02     NaN  ...    0.024  0.036    NaN    NaN  0.048  0.036  0.048   \n",
       "2010-12-03     NaN  ...    0.024    NaN  0.049  0.049  0.073  0.024  0.049   \n",
       "2010-12-05   0.031  ...    0.108  0.031  0.046  0.046  0.046  0.031    NaN   \n",
       "2010-12-06   0.083  ...    0.056  0.028  0.069  0.042    NaN    NaN    NaN   \n",
       "2010-12-07     NaN  ...    0.065  0.130  0.065    NaN    NaN    NaN    NaN   \n",
       "2010-12-08   0.014  ...    0.041  0.054  0.027    NaN    NaN    NaN    NaN   \n",
       "2010-12-09     NaN  ...    0.059  0.015    NaN    NaN    NaN    NaN    NaN   \n",
       "2010-12-10     NaN  ...      NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "2010-12-12     NaN  ...      NaN    NaN    NaN    NaN    NaN    NaN    NaN   \n",
       "\n",
       "CohortIndex    371    372    373  \n",
       "CohortDay                         \n",
       "2010-12-01   0.085  0.061  0.049  \n",
       "2010-12-02   0.024  0.012    NaN  \n",
       "2010-12-03     NaN    NaN    NaN  \n",
       "2010-12-05     NaN    NaN    NaN  \n",
       "2010-12-06     NaN    NaN    NaN  \n",
       "2010-12-07     NaN    NaN    NaN  \n",
       "2010-12-08     NaN    NaN    NaN  \n",
       "2010-12-09     NaN    NaN    NaN  \n",
       "2010-12-10     NaN    NaN    NaN  \n",
       "2010-12-12     NaN    NaN    NaN  \n",
       "\n",
       "[10 rows x 374 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# get the proportional values of the numbers per row\n",
    "cohort_sizes = cohort_wide.iloc[:, 0]\n",
    "retention = cohort_wide.divide(cohort_sizes, axis = 0)\n",
    "\n",
    "retention.round(3).head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1-2. Average Price"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>CohortDay</th>\n",
       "      <th>CohortIndex</th>\n",
       "      <th>UnitPrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>0</td>\n",
       "      <td>3.526332</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>1</td>\n",
       "      <td>4.094222</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>2</td>\n",
       "      <td>3.433913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>4</td>\n",
       "      <td>4.250233</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>5</td>\n",
       "      <td>3.344545</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CohortDay  CohortIndex  UnitPrice\n",
       "0 2010-12-01            0   3.526332\n",
       "1 2010-12-01            1   4.094222\n",
       "2 2010-12-01            2   3.433913\n",
       "3 2010-12-01            4   4.250233\n",
       "4 2010-12-01            5   3.344545"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cohort_data2 = online.groupby(['CohortDay', 'CohortIndex'])['UnitPrice'].mean().reset_index()\n",
    "cohort_data2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>CohortIndex</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>364</th>\n",
       "      <th>365</th>\n",
       "      <th>366</th>\n",
       "      <th>367</th>\n",
       "      <th>368</th>\n",
       "      <th>369</th>\n",
       "      <th>370</th>\n",
       "      <th>371</th>\n",
       "      <th>372</th>\n",
       "      <th>373</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CohortDay</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",
       "      <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",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2010-12-01</th>\n",
       "      <td>3.5</td>\n",
       "      <td>4.1</td>\n",
       "      <td>3.4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.3</td>\n",
       "      <td>3.3</td>\n",
       "      <td>2.6</td>\n",
       "      <td>3.2</td>\n",
       "      <td>3.1</td>\n",
       "      <td>4.6</td>\n",
       "      <td>...</td>\n",
       "      <td>3.4</td>\n",
       "      <td>2.9</td>\n",
       "      <td>4.2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.9</td>\n",
       "      <td>1.9</td>\n",
       "      <td>3.3</td>\n",
       "      <td>2.4</td>\n",
       "      <td>2.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-02</th>\n",
       "      <td>2.9</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.4</td>\n",
       "      <td>1.8</td>\n",
       "      <td>2.9</td>\n",
       "      <td>4.4</td>\n",
       "      <td>2.9</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>4.6</td>\n",
       "      <td>3.9</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.3</td>\n",
       "      <td>3.7</td>\n",
       "      <td>2.9</td>\n",
       "      <td>2.4</td>\n",
       "      <td>2.5</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-03</th>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.8</td>\n",
       "      <td>2.6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.1</td>\n",
       "      <td>4.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>2.7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.9</td>\n",
       "      <td>1.8</td>\n",
       "      <td>1.9</td>\n",
       "      <td>3.6</td>\n",
       "      <td>2.4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-05</th>\n",
       "      <td>2.9</td>\n",
       "      <td>3.7</td>\n",
       "      <td>5.3</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.7</td>\n",
       "      <td>2.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.4</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.8</td>\n",
       "      <td>...</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.8</td>\n",
       "      <td>3.4</td>\n",
       "      <td>3.1</td>\n",
       "      <td>2.6</td>\n",
       "      <td>2.9</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-06</th>\n",
       "      <td>2.9</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.5</td>\n",
       "      <td>1.2</td>\n",
       "      <td>7.4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.4</td>\n",
       "      <td>4.5</td>\n",
       "      <td>3.4</td>\n",
       "      <td>3.6</td>\n",
       "      <td>...</td>\n",
       "      <td>3.9</td>\n",
       "      <td>1.2</td>\n",
       "      <td>3.5</td>\n",
       "      <td>3.9</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-07</th>\n",
       "      <td>3.3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.8</td>\n",
       "      <td>3.8</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.1</td>\n",
       "      <td>2.7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>1.9</td>\n",
       "      <td>1.8</td>\n",
       "      <td>4.7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-08</th>\n",
       "      <td>3.4</td>\n",
       "      <td>2.1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.4</td>\n",
       "      <td>2.4</td>\n",
       "      <td>2.5</td>\n",
       "      <td>3.8</td>\n",
       "      <td>...</td>\n",
       "      <td>2.7</td>\n",
       "      <td>2.5</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-09</th>\n",
       "      <td>3.1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.2</td>\n",
       "      <td>2.3</td>\n",
       "      <td>3.6</td>\n",
       "      <td>3.7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>1.6</td>\n",
       "      <td>4.6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-10</th>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.4</td>\n",
       "      <td>7.4</td>\n",
       "      <td>4.5</td>\n",
       "      <td>4.1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010-12-12</th>\n",
       "      <td>2.4</td>\n",
       "      <td>5.4</td>\n",
       "      <td>2.8</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.5</td>\n",
       "      <td>2.3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 374 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "CohortIndex  0    1    2    3    4    5    6    7    8    9   ...   364  365  \\\n",
       "CohortDay                                                     ...              \n",
       "2010-12-01   3.5  4.1  3.4  NaN  4.3  3.3  2.6  3.2  3.1  4.6 ...   3.4  2.9   \n",
       "2010-12-02   2.9  NaN  NaN  NaN  3.4  1.8  2.9  4.4  2.9  NaN ...   4.6  3.9   \n",
       "2010-12-03   3.0  NaN  1.8  2.6  NaN  3.1  4.5  NaN  NaN  NaN ...   2.7  NaN   \n",
       "2010-12-05   2.9  3.7  5.3  2.0  3.7  2.5  NaN  3.4  3.0  4.8 ...   3.0  1.8   \n",
       "2010-12-06   2.9  3.0  1.5  1.2  7.4  NaN  1.4  4.5  3.4  3.6 ...   3.9  1.2   \n",
       "2010-12-07   3.3  NaN  2.8  3.8  NaN  NaN  3.1  2.7  NaN  NaN ...   1.9  1.8   \n",
       "2010-12-08   3.4  2.1  NaN  NaN  1.0  1.0  4.4  2.4  2.5  3.8 ...   2.7  2.5   \n",
       "2010-12-09   3.1  NaN  NaN  NaN  3.0  2.2  2.3  3.6  3.7  NaN ...   1.6  4.6   \n",
       "2010-12-10   3.0  NaN  NaN  2.0  2.4  7.4  4.5  4.1  NaN  NaN ...   NaN  NaN   \n",
       "2010-12-12   2.4  5.4  2.8  NaN  NaN  NaN  NaN  4.5  2.3  NaN ...   NaN  NaN   \n",
       "\n",
       "CohortIndex  366  367  368  369  370  371  372  373  \n",
       "CohortDay                                            \n",
       "2010-12-01   4.2  NaN  2.0  1.9  1.9  3.3  2.4  2.9  \n",
       "2010-12-02   NaN  NaN  3.3  3.7  2.9  2.4  2.5  NaN  \n",
       "2010-12-03   1.9  1.8  1.9  3.6  2.4  NaN  NaN  NaN  \n",
       "2010-12-05   3.4  3.1  2.6  2.9  NaN  NaN  NaN  NaN  \n",
       "2010-12-06   3.5  3.9  NaN  NaN  NaN  NaN  NaN  NaN  \n",
       "2010-12-07   4.7  NaN  NaN  NaN  NaN  NaN  NaN  NaN  \n",
       "2010-12-08   2.0  NaN  NaN  NaN  NaN  NaN  NaN  NaN  \n",
       "2010-12-09   NaN  NaN  NaN  NaN  NaN  NaN  NaN  NaN  \n",
       "2010-12-10   NaN  NaN  NaN  NaN  NaN  NaN  NaN  NaN  \n",
       "2010-12-12   NaN  NaN  NaN  NaN  NaN  NaN  NaN  NaN  \n",
       "\n",
       "[10 rows x 374 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# pivot the table \n",
    "cohort_wide2 = cohort_data2.pivot(index= 'CohortDay', columns= 'CohortIndex', values= 'UnitPrice').round(1)\n",
    "cohort_wide2.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2) RFM Segmentation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "RFM (Recency, Frequency, Monetary) is a very Simple Technique that we can apply it very easy and get the super useful analysis for our Customer Segmentation. Recency is days since the customers made the last purchase and by definition, the lower it is the better. Frequency is the number of transaction in the last 12 months. Monetary value is the total amout of money the customers spent in the last 12 months."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>InvoiceNo</th>\n",
       "      <th>StockCode</th>\n",
       "      <th>Description</th>\n",
       "      <th>Quantity</th>\n",
       "      <th>InvoiceDate</th>\n",
       "      <th>UnitPrice</th>\n",
       "      <th>CustomerID</th>\n",
       "      <th>Country</th>\n",
       "      <th>InvoiceDay</th>\n",
       "      <th>CohortDay</th>\n",
       "      <th>CohortIndex</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>548670</td>\n",
       "      <td>21975</td>\n",
       "      <td>PACK OF 60 DINOSAUR CAKE CASES</td>\n",
       "      <td>1</td>\n",
       "      <td>2011-04-01 15:37:00</td>\n",
       "      <td>0.55</td>\n",
       "      <td>15356.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-04-01</td>\n",
       "      <td>2010-12-06</td>\n",
       "      <td>120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>580513</td>\n",
       "      <td>23321</td>\n",
       "      <td>SMALL WHITE HEART OF WICKER</td>\n",
       "      <td>2</td>\n",
       "      <td>2011-12-04 13:59:00</td>\n",
       "      <td>1.65</td>\n",
       "      <td>14456.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-12-04</td>\n",
       "      <td>2011-07-20</td>\n",
       "      <td>134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>568531</td>\n",
       "      <td>23505</td>\n",
       "      <td>PLAYING CARDS I LOVE LONDON</td>\n",
       "      <td>3</td>\n",
       "      <td>2011-09-27 13:49:00</td>\n",
       "      <td>1.25</td>\n",
       "      <td>16713.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-09-27</td>\n",
       "      <td>2010-12-08</td>\n",
       "      <td>294</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>544178</td>\n",
       "      <td>21411</td>\n",
       "      <td>GINGHAM HEART  DOORSTOP RED</td>\n",
       "      <td>3</td>\n",
       "      <td>2011-02-16 14:40:00</td>\n",
       "      <td>4.25</td>\n",
       "      <td>14543.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-02-16</td>\n",
       "      <td>2010-12-10</td>\n",
       "      <td>71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>565381</td>\n",
       "      <td>22457</td>\n",
       "      <td>NATURAL SLATE HEART CHALKBOARD</td>\n",
       "      <td>6</td>\n",
       "      <td>2011-09-02 15:23:00</td>\n",
       "      <td>2.95</td>\n",
       "      <td>16173.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-09-02</td>\n",
       "      <td>2011-09-02</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  InvoiceNo StockCode                      Description  Quantity  \\\n",
       "0    548670     21975   PACK OF 60 DINOSAUR CAKE CASES         1   \n",
       "1    580513     23321      SMALL WHITE HEART OF WICKER         2   \n",
       "2    568531     23505     PLAYING CARDS I LOVE LONDON          3   \n",
       "3    544178     21411      GINGHAM HEART  DOORSTOP RED         3   \n",
       "4    565381     22457  NATURAL SLATE HEART CHALKBOARD          6   \n",
       "\n",
       "          InvoiceDate  UnitPrice  CustomerID         Country InvoiceDay  \\\n",
       "0 2011-04-01 15:37:00       0.55     15356.0  United Kingdom 2011-04-01   \n",
       "1 2011-12-04 13:59:00       1.65     14456.0  United Kingdom 2011-12-04   \n",
       "2 2011-09-27 13:49:00       1.25     16713.0  United Kingdom 2011-09-27   \n",
       "3 2011-02-16 14:40:00       4.25     14543.0  United Kingdom 2011-02-16   \n",
       "4 2011-09-02 15:23:00       2.95     16173.0  United Kingdom 2011-09-02   \n",
       "\n",
       "   CohortDay  CohortIndex  \n",
       "0 2010-12-06          120  \n",
       "1 2011-07-20          134  \n",
       "2 2010-12-08          294  \n",
       "3 2010-12-10           71  \n",
       "4 2011-09-02            0  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "online.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's create a new dataframe with RFM values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead 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>InvoiceNo</th>\n",
       "      <th>StockCode</th>\n",
       "      <th>Description</th>\n",
       "      <th>Quantity</th>\n",
       "      <th>InvoiceDate</th>\n",
       "      <th>UnitPrice</th>\n",
       "      <th>CustomerID</th>\n",
       "      <th>Country</th>\n",
       "      <th>InvoiceDay</th>\n",
       "      <th>CohortDay</th>\n",
       "      <th>CohortIndex</th>\n",
       "      <th>TotalSum</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>548670</td>\n",
       "      <td>21975</td>\n",
       "      <td>PACK OF 60 DINOSAUR CAKE CASES</td>\n",
       "      <td>1</td>\n",
       "      <td>2011-04-01 15:37:00</td>\n",
       "      <td>0.55</td>\n",
       "      <td>15356.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-04-01</td>\n",
       "      <td>2010-12-06</td>\n",
       "      <td>120</td>\n",
       "      <td>0.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>580513</td>\n",
       "      <td>23321</td>\n",
       "      <td>SMALL WHITE HEART OF WICKER</td>\n",
       "      <td>2</td>\n",
       "      <td>2011-12-04 13:59:00</td>\n",
       "      <td>1.65</td>\n",
       "      <td>14456.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-12-04</td>\n",
       "      <td>2011-07-20</td>\n",
       "      <td>134</td>\n",
       "      <td>3.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>568531</td>\n",
       "      <td>23505</td>\n",
       "      <td>PLAYING CARDS I LOVE LONDON</td>\n",
       "      <td>3</td>\n",
       "      <td>2011-09-27 13:49:00</td>\n",
       "      <td>1.25</td>\n",
       "      <td>16713.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-09-27</td>\n",
       "      <td>2010-12-08</td>\n",
       "      <td>294</td>\n",
       "      <td>3.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>544178</td>\n",
       "      <td>21411</td>\n",
       "      <td>GINGHAM HEART  DOORSTOP RED</td>\n",
       "      <td>3</td>\n",
       "      <td>2011-02-16 14:40:00</td>\n",
       "      <td>4.25</td>\n",
       "      <td>14543.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-02-16</td>\n",
       "      <td>2010-12-10</td>\n",
       "      <td>71</td>\n",
       "      <td>12.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>565381</td>\n",
       "      <td>22457</td>\n",
       "      <td>NATURAL SLATE HEART CHALKBOARD</td>\n",
       "      <td>6</td>\n",
       "      <td>2011-09-02 15:23:00</td>\n",
       "      <td>2.95</td>\n",
       "      <td>16173.0</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>2011-09-02</td>\n",
       "      <td>2011-09-02</td>\n",
       "      <td>0</td>\n",
       "      <td>17.70</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  InvoiceNo StockCode                      Description  Quantity  \\\n",
       "0    548670     21975   PACK OF 60 DINOSAUR CAKE CASES         1   \n",
       "1    580513     23321      SMALL WHITE HEART OF WICKER         2   \n",
       "2    568531     23505     PLAYING CARDS I LOVE LONDON          3   \n",
       "3    544178     21411      GINGHAM HEART  DOORSTOP RED         3   \n",
       "4    565381     22457  NATURAL SLATE HEART CHALKBOARD          6   \n",
       "\n",
       "          InvoiceDate  UnitPrice  CustomerID         Country InvoiceDay  \\\n",
       "0 2011-04-01 15:37:00       0.55     15356.0  United Kingdom 2011-04-01   \n",
       "1 2011-12-04 13:59:00       1.65     14456.0  United Kingdom 2011-12-04   \n",
       "2 2011-09-27 13:49:00       1.25     16713.0  United Kingdom 2011-09-27   \n",
       "3 2011-02-16 14:40:00       4.25     14543.0  United Kingdom 2011-02-16   \n",
       "4 2011-09-02 15:23:00       2.95     16173.0  United Kingdom 2011-09-02   \n",
       "\n",
       "   CohortDay  CohortIndex  TotalSum  \n",
       "0 2010-12-06          120      0.55  \n",
       "1 2011-07-20          134      3.30  \n",
       "2 2010-12-08          294      3.75  \n",
       "3 2010-12-10           71     12.75  \n",
       "4 2011-09-02            0     17.70  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create total spend dataframe\n",
    "online['TotalSum'] = online.Quantity * online.UnitPrice\n",
    "\n",
    "online.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Min : 2010-12-01 00:00:00, Max : 2011-12-09 00:00:00\n"
     ]
    }
   ],
   "source": [
    "print('Min : {}, Max : {}'.format(min(online.InvoiceDay), max(online.InvoiceDay)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The last day of purchase in total is 09 DEC, 2011. To calculate the day periods, let's set one day after the last one, or 10 DEC as a snapshot date. We will cound the diff days with snapshot_date."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "snapshot_date = max(online.InvoiceDay) + dt.timedelta(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# calculate RFM values\n",
    "datamart = online.groupby('CustomerID').agg({\n",
    "    'InvoiceDate' : lambda x: (snapshot_date - x.max()).days,\n",
    "    'InvoiceNo' : 'count', \n",
    "    'TotalSum' : 'sum'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>MonetaryValue</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CustomerID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12347.0</th>\n",
       "      <td>2</td>\n",
       "      <td>29</td>\n",
       "      <td>641.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12348.0</th>\n",
       "      <td>248</td>\n",
       "      <td>3</td>\n",
       "      <td>400.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12349.0</th>\n",
       "      <td>18</td>\n",
       "      <td>16</td>\n",
       "      <td>324.42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12350.0</th>\n",
       "      <td>310</td>\n",
       "      <td>5</td>\n",
       "      <td>110.40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12352.0</th>\n",
       "      <td>36</td>\n",
       "      <td>17</td>\n",
       "      <td>621.28</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Recency  Frequency  MonetaryValue\n",
       "CustomerID                                   \n",
       "12347.0           2         29         641.00\n",
       "12348.0         248          3         400.00\n",
       "12349.0          18         16         324.42\n",
       "12350.0         310          5         110.40\n",
       "12352.0          36         17         621.28"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# renmae the columns\n",
    "datamart.rename(columns = {'InvoiceDate' : 'Recency', \n",
    "                           'InvoiceNo' : 'Frequency', \n",
    "                           'TotalSum' : 'MonetaryValue'}, inplace = True)\n",
    "\n",
    "datamart.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As the three columns are grouped by customers and count the days from the max date value, `Recency` is the days since the last purchase of a customer. `Frequency` is the number of purchases of a customer and `MonetaryValue` is the total amount of spend of a customer.  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2-1. RFM quartiles "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's group the customers based on `Recency` and `Frequency`. We will use quantile values to get three equal percentile groups an then make three separate gruops. As the lower `Recency` value is the better, we will label them in decreasing order.      "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CustomerID\n",
       "12347.0    4\n",
       "12348.0    4\n",
       "12349.0    3\n",
       "12350.0    2\n",
       "12352.0    4\n",
       "Name: MonetaryValue, dtype: category\n",
       "Categories (4, int64): [1 < 2 < 3 < 4]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create labels and assign them to tree percntile groups \n",
    "r_labels = range(4, 0, -1)\n",
    "r_groups = pd.qcut(datamart.Recency, q = 4, labels = r_labels)\n",
    "\n",
    "f_labels = range(1, 5)\n",
    "f_groups = pd.qcut(datamart.Frequency, q = 4, labels = f_labels)\n",
    "\n",
    "m_labels = range(1, 5)\n",
    "m_groups = pd.qcut(datamart.MonetaryValue, q = 4, labels = m_labels)\n",
    "\n",
    "m_groups.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# make a new column for group labels\n",
    "datamart['R'] = r_groups.values\n",
    "datamart['F'] = f_groups.values\n",
    "datamart['M'] = m_groups.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>MonetaryValue</th>\n",
       "      <th>R</th>\n",
       "      <th>F</th>\n",
       "      <th>M</th>\n",
       "      <th>RFM_Segment</th>\n",
       "      <th>RFM_Score</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CustomerID</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>12347.0</th>\n",
       "      <td>2</td>\n",
       "      <td>29</td>\n",
       "      <td>641.00</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>444</td>\n",
       "      <td>12.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12348.0</th>\n",
       "      <td>248</td>\n",
       "      <td>3</td>\n",
       "      <td>400.00</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>114</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12349.0</th>\n",
       "      <td>18</td>\n",
       "      <td>16</td>\n",
       "      <td>324.42</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>433</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12350.0</th>\n",
       "      <td>310</td>\n",
       "      <td>5</td>\n",
       "      <td>110.40</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>122</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12352.0</th>\n",
       "      <td>36</td>\n",
       "      <td>17</td>\n",
       "      <td>621.28</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>334</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Recency  Frequency  MonetaryValue  R  F  M RFM_Segment  RFM_Score\n",
       "CustomerID                                                                   \n",
       "12347.0           2         29         641.00  4  4  4         444       12.0\n",
       "12348.0         248          3         400.00  1  1  4         114        6.0\n",
       "12349.0          18         16         324.42  4  3  3         433       10.0\n",
       "12350.0         310          5         110.40  1  2  2         122        5.0\n",
       "12352.0          36         17         621.28  3  3  4         334       10.0"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# sum up the three columns\n",
    "datamart['RFM_Segment'] = datamart.apply(lambda x: str(x['R']) + str(x['F']) + str(x['M']), axis = 1)\n",
    "datamart['RFM_Score'] = datamart[['R', 'F', 'M']].sum(axis = 1)\n",
    "datamart.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2-2. RFM Segment Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "With this value, we can go further analysis such as what is the average values for each RFM values or leveling\n",
    "customers in total RFM score."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>Recency</th>\n",
       "      <th>Frequency</th>\n",
       "      <th colspan=\"2\" halign=\"left\">MonetaryValue</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>mean</th>\n",
       "      <th>mean</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RFM_Score</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3.0</th>\n",
       "      <td>264.3</td>\n",
       "      <td>2.1</td>\n",
       "      <td>21.4</td>\n",
       "      <td>386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4.0</th>\n",
       "      <td>173.4</td>\n",
       "      <td>3.0</td>\n",
       "      <td>44.2</td>\n",
       "      <td>384</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5.0</th>\n",
       "      <td>142.5</td>\n",
       "      <td>4.7</td>\n",
       "      <td>75.8</td>\n",
       "      <td>487</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6.0</th>\n",
       "      <td>100.1</td>\n",
       "      <td>6.2</td>\n",
       "      <td>212.2</td>\n",
       "      <td>438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7.0</th>\n",
       "      <td>82.8</td>\n",
       "      <td>9.1</td>\n",
       "      <td>168.4</td>\n",
       "      <td>429</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Recency Frequency MonetaryValue      \n",
       "             mean      mean          mean count\n",
       "RFM_Score                                      \n",
       "3.0         264.3       2.1          21.4   386\n",
       "4.0         173.4       3.0          44.2   384\n",
       "5.0         142.5       4.7          75.8   487\n",
       "6.0         100.1       6.2         212.2   438\n",
       "7.0          82.8       9.1         168.4   429"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# calculate averae values for each RFM_score\n",
    "datamart_agg = datamart.groupby('RFM_Score').agg({\n",
    "    'Recency' : 'mean',\n",
    "    'Frequency' : 'mean',\n",
    "    'MonetaryValue' : ['mean', 'count']\n",
    "})\n",
    "\n",
    "datamart_agg.round(1).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define rfm_level function\n",
    "def rfm_level(df):\n",
    "    if df['RFM_Score'] >= 10:\n",
    "        return 'Top'\n",
    "    elif (df['RFM_Score'] >= 6) and (df['RFM_Score'] < 10):\n",
    "        return 'Mid'\n",
    "    else:\n",
    "        return 'Low'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>MonetaryValue</th>\n",
       "      <th>R</th>\n",
       "      <th>F</th>\n",
       "      <th>M</th>\n",
       "      <th>RFM_Segment</th>\n",
       "      <th>RFM_Score</th>\n",
       "      <th>RFM_Level</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CustomerID</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>12347.0</th>\n",
       "      <td>2</td>\n",
       "      <td>29</td>\n",
       "      <td>641.00</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>444</td>\n",
       "      <td>12.0</td>\n",
       "      <td>Top</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12348.0</th>\n",
       "      <td>248</td>\n",
       "      <td>3</td>\n",
       "      <td>400.00</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>114</td>\n",
       "      <td>6.0</td>\n",
       "      <td>Mid</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12349.0</th>\n",
       "      <td>18</td>\n",
       "      <td>16</td>\n",
       "      <td>324.42</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>433</td>\n",
       "      <td>10.0</td>\n",
       "      <td>Top</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12350.0</th>\n",
       "      <td>310</td>\n",
       "      <td>5</td>\n",
       "      <td>110.40</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>122</td>\n",
       "      <td>5.0</td>\n",
       "      <td>Low</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12352.0</th>\n",
       "      <td>36</td>\n",
       "      <td>17</td>\n",
       "      <td>621.28</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>334</td>\n",
       "      <td>10.0</td>\n",
       "      <td>Top</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Recency  Frequency  MonetaryValue  R  F  M RFM_Segment  RFM_Score  \\\n",
       "CustomerID                                                                      \n",
       "12347.0           2         29         641.00  4  4  4         444       12.0   \n",
       "12348.0         248          3         400.00  1  1  4         114        6.0   \n",
       "12349.0          18         16         324.42  4  3  3         433       10.0   \n",
       "12350.0         310          5         110.40  1  2  2         122        5.0   \n",
       "12352.0          36         17         621.28  3  3  4         334       10.0   \n",
       "\n",
       "           RFM_Level  \n",
       "CustomerID            \n",
       "12347.0          Top  \n",
       "12348.0          Mid  \n",
       "12349.0          Top  \n",
       "12350.0          Low  \n",
       "12352.0          Top  "
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create a new variable RFM_Level\n",
    "datamart['RFM_Level'] = datamart.apply(rfm_level, axis=1)\n",
    "datamart.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3) Customer Segmentation with Kmeans"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can also apply Kmeans clustering with RFM values. As Kmeans clustering require data to be normalized and has a symmetric distribution, preprocessing process in scale is needed. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3-1. Preprocessing "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\jjone\\Anaconda3\\lib\\site-packages\\scipy\\stats\\stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the distribution of RFM values\n",
    "plt.subplot(3, 1, 1); sns.distplot(datamart.Recency)\n",
    "plt.subplot(3, 1, 2); sns.distplot(datamart.Frequency)\n",
    "plt.subplot(3, 1, 3); sns.distplot(datamart.MonetaryValue)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can see above, the values are skewed and need to be normalized. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Recency</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>MonetaryValue</th>\n",
       "      <th>RFM_Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4121.000000</td>\n",
       "      <td>4121.000000</td>\n",
       "      <td>4121.000000</td>\n",
       "      <td>4121.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>94.147780</td>\n",
       "      <td>19.744237</td>\n",
       "      <td>455.239765</td>\n",
       "      <td>7.439456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>100.962757</td>\n",
       "      <td>47.947383</td>\n",
       "      <td>3199.513146</td>\n",
       "      <td>2.825471</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-817.510000</td>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>18.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>57.500000</td>\n",
       "      <td>5.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>51.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>137.160000</td>\n",
       "      <td>7.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>149.000000</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>340.930000</td>\n",
       "      <td>10.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>373.000000</td>\n",
       "      <td>1634.000000</td>\n",
       "      <td>168471.250000</td>\n",
       "      <td>12.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Recency    Frequency  MonetaryValue    RFM_Score\n",
       "count  4121.000000  4121.000000    4121.000000  4121.000000\n",
       "mean     94.147780    19.744237     455.239765     7.439456\n",
       "std     100.962757    47.947383    3199.513146     2.825471\n",
       "min       0.000000     1.000000    -817.510000     3.000000\n",
       "25%      18.000000     4.000000      57.500000     5.000000\n",
       "50%      51.000000     9.000000     137.160000     7.000000\n",
       "75%     149.000000    22.000000     340.930000    10.000000\n",
       "max     373.000000  1634.000000  168471.250000    12.000000"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# summary values\n",
    "datamart.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Due to the zero or negative values in `Recency` and `MonetaryValue`, we need to set them 1 before log transformation and scaling."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# define function for the values below 0\n",
    "def neg_to_zero(x):\n",
    "    if x <= 0:\n",
    "        return 1\n",
    "    else:\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>MonetaryValue</th>\n",
       "      <th>R</th>\n",
       "      <th>F</th>\n",
       "      <th>M</th>\n",
       "      <th>RFM_Segment</th>\n",
       "      <th>RFM_Score</th>\n",
       "      <th>RFM_Level</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CustomerID</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>12347.0</th>\n",
       "      <td>2</td>\n",
       "      <td>29</td>\n",
       "      <td>641.00</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>444</td>\n",
       "      <td>12.0</td>\n",
       "      <td>Top</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12348.0</th>\n",
       "      <td>248</td>\n",
       "      <td>3</td>\n",
       "      <td>400.00</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>114</td>\n",
       "      <td>6.0</td>\n",
       "      <td>Mid</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12349.0</th>\n",
       "      <td>18</td>\n",
       "      <td>16</td>\n",
       "      <td>324.42</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>433</td>\n",
       "      <td>10.0</td>\n",
       "      <td>Top</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12350.0</th>\n",
       "      <td>310</td>\n",
       "      <td>5</td>\n",
       "      <td>110.40</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>122</td>\n",
       "      <td>5.0</td>\n",
       "      <td>Low</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12352.0</th>\n",
       "      <td>36</td>\n",
       "      <td>17</td>\n",
       "      <td>621.28</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>334</td>\n",
       "      <td>10.0</td>\n",
       "      <td>Top</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Recency  Frequency  MonetaryValue  R  F  M RFM_Segment  RFM_Score  \\\n",
       "CustomerID                                                                      \n",
       "12347.0           2         29         641.00  4  4  4         444       12.0   \n",
       "12348.0         248          3         400.00  1  1  4         114        6.0   \n",
       "12349.0          18         16         324.42  4  3  3         433       10.0   \n",
       "12350.0         310          5         110.40  1  2  2         122        5.0   \n",
       "12352.0          36         17         621.28  3  3  4         334       10.0   \n",
       "\n",
       "           RFM_Level  \n",
       "CustomerID            \n",
       "12347.0          Top  \n",
       "12348.0          Mid  \n",
       "12349.0          Top  \n",
       "12350.0          Low  \n",
       "12352.0          Top  "
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# apply the function to Recency and MonetaryValue column \n",
    "datamart['Recency'] = [neg_to_zero(x) for x in datamart.Recency]\n",
    "datamart['MonetaryValue'] = [neg_to_zero(x) for x in datamart.MonetaryValue]\n",
    "datamart.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "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>MonetaryValue</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CustomerID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12347.0</th>\n",
       "      <td>0.693</td>\n",
       "      <td>3.367</td>\n",
       "      <td>6.463</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12348.0</th>\n",
       "      <td>5.513</td>\n",
       "      <td>1.099</td>\n",
       "      <td>5.991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12349.0</th>\n",
       "      <td>2.890</td>\n",
       "      <td>2.773</td>\n",
       "      <td>5.782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12350.0</th>\n",
       "      <td>5.737</td>\n",
       "      <td>1.609</td>\n",
       "      <td>4.704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12352.0</th>\n",
       "      <td>3.584</td>\n",
       "      <td>2.833</td>\n",
       "      <td>6.432</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Recency  Frequency  MonetaryValue\n",
       "CustomerID                                   \n",
       "12347.0       0.693      3.367          6.463\n",
       "12348.0       5.513      1.099          5.991\n",
       "12349.0       2.890      2.773          5.782\n",
       "12350.0       5.737      1.609          4.704\n",
       "12352.0       3.584      2.833          6.432"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# unskew the data\n",
    "rfm_log = datamart[['Recency', 'Frequency', 'MonetaryValue']].apply(np.log, axis = 1).round(3)\n",
    "rfm_log.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>MonetaryValue</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4121.000000</td>\n",
       "      <td>4121.000000</td>\n",
       "      <td>4121.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>3.772734</td>\n",
       "      <td>2.213949</td>\n",
       "      <td>4.921445</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.463306</td>\n",
       "      <td>1.211813</td>\n",
       "      <td>1.420750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-0.799000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2.890000</td>\n",
       "      <td>1.386000</td>\n",
       "      <td>4.052000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>3.932000</td>\n",
       "      <td>2.197000</td>\n",
       "      <td>4.921000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>5.004000</td>\n",
       "      <td>3.091000</td>\n",
       "      <td>5.832000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>5.922000</td>\n",
       "      <td>7.399000</td>\n",
       "      <td>12.035000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Recency    Frequency  MonetaryValue\n",
       "count  4121.000000  4121.000000    4121.000000\n",
       "mean      3.772734     2.213949       4.921445\n",
       "std       1.463306     1.211813       1.420750\n",
       "min       0.000000     0.000000      -0.799000\n",
       "25%       2.890000     1.386000       4.052000\n",
       "50%       3.932000     2.197000       4.921000\n",
       "75%       5.004000     3.091000       5.832000\n",
       "max       5.922000     7.399000      12.035000"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfm_log.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# scale the data\n",
    "scaler = StandardScaler()\n",
    "rfm_scaled = scaler.fit_transform(rfm_log)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "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>MonetaryValue</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CustomerID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12347.0</th>\n",
       "      <td>-2.104896</td>\n",
       "      <td>0.951625</td>\n",
       "      <td>1.085161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12348.0</th>\n",
       "      <td>1.189415</td>\n",
       "      <td>-0.920178</td>\n",
       "      <td>0.752902</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12349.0</th>\n",
       "      <td>-0.603319</td>\n",
       "      <td>0.461391</td>\n",
       "      <td>0.605778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12350.0</th>\n",
       "      <td>1.342511</td>\n",
       "      <td>-0.499270</td>\n",
       "      <td>-0.153068</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12352.0</th>\n",
       "      <td>-0.128993</td>\n",
       "      <td>0.510909</td>\n",
       "      <td>1.063339</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Recency  Frequency  MonetaryValue\n",
       "CustomerID                                    \n",
       "12347.0    -2.104896   0.951625       1.085161\n",
       "12348.0     1.189415  -0.920178       0.752902\n",
       "12349.0    -0.603319   0.461391       0.605778\n",
       "12350.0     1.342511  -0.499270      -0.153068\n",
       "12352.0    -0.128993   0.510909       1.063339"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# transform into a dataframe\n",
    "rfm_scaled = pd.DataFrame(rfm_scaled, index = datamart.index, columns = rfm_log.columns)\n",
    "rfm_scaled.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\jjone\\Anaconda3\\lib\\site-packages\\scipy\\stats\\stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the distribution of RFM values\n",
    "plt.subplot(3, 1, 1); sns.distplot(rfm_scaled.Recency)\n",
    "plt.subplot(3, 1, 2); sns.distplot(rfm_scaled.Frequency)\n",
    "plt.subplot(3, 1, 3); sns.distplot(rfm_scaled.MonetaryValue)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3-2. K-means clustering"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "With the Elbow method, we can get the optimal number of clusters.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "# initiate an empty dictionary\n",
    "wcss = {}\n",
    "\n",
    "# Elbow method with for loop\n",
    "for i in range(1, 21):\n",
    "    kmeans = KMeans(n_clusters= i, init= 'k-means++', max_iter= 300)\n",
    "    kmeans.fit(rfm_scaled)\n",
    "    wcss[i] = kmeans.inertia_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAD8CAYAAACcjGjIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzt3Xl8XHW9//HXJ/vaJG3TdEnbdEkLZW0pBQTL0ooFkYIKgigV0HoVFOT6u8LlXjfEiyKyqHAv+6JsIpuAQAURBOm+7+meNm2zNs3SrN/fH3PSTjpLMpM0STPv5+MxjznzPef7ne+kp/OZ73K+x5xziIiI+Ivr7QqIiEjfo+AgIiIBFBxERCSAgoOIiARQcBARkQAKDiIiEkDBQUREAig4iIhIAAUHEREJkNDbFYjW4MGDXUFBQW9XQ0TkqLJ48eIy51xuR8cdtcGhoKCARYsW9XY1RESOKma2rTPHqVtJREQCKDiIiEgABQcREQmg4CAiIgEUHEREJICCg4iIBFBwEBGRAEftdQ7B7Kqq5/GPtvDeur20tDpOHzuI684aQ2FeZm9XTUTkqNJvgsPakmqufOgTquqbDqZtLa/jpaU7efjqqZw9ocMLAkVExNMvupWcc/zgT8vbBYY2jc2t3Pz8Mg40tfRCzUREjk79Ijis3lXN6l3VIfeX1zby7tq9PVgjEZGjW78IDsWV9R0es7OqrgdqIiLSP/SL4DA8O6XDY4ZlpfZATURE+od+ERxOGJHFMUNDz0jKSUvkM5PyerBGIiJHtw6Dg5k9ZmZ7zWyVX9pdZrbOzFaY2ctmlu2371YzKzKz9Wb2Wb/0WV5akZnd4pc+xszmm9lGM3vezJIi/RBmxl1fOonMlMDJV/EGd33pJFIS4yMtVkQkZnWm5fAEMOuwtHnA8c65E4ENwK0AZjYJuAI4zsvzgJnFm1k88HvgAmAScKV3LMAvgXucc4VAJXBdNB/khPws3vjup5lzxmjSkg4FgieuncZMtRpERCLSYXBwzn0AVByW9o5zrtl7+QmQ723PBp5zzjU457YARcA071HknNvsnGsEngNmm5kB5wEvevmfBC6J9sOMGpTGT2cfz/Xnjj/0Ac2iLU5EJGZ1x5jDtcBfve0RwA6/fcVeWqj0QUCVX6BpS++S8UMyDm5v3LO/q8WJiMScLgUHM7sNaAb+2JYU5DAXRXqo95trZovMbFFpaWnIek3wWy5jw96akMeJiEhwUQcHM5sDXARc5Zxr+0IvBkb6HZYP7AqTXgZkm1nCYelBOececs5Ndc5Nzc0NvRzGyJxUkhJ8H61oj4KDiEikogoOZjYL+CFwsXPO/+qy14ArzCzZzMYAhcACYCFQ6M1MSsI3aP2aF1T+DnzJyz8HeDW6j3JIQnwcYwenA7Bh734OxS4REemMzkxlfRb4FzDRzIrN7Drgd0AmMM/MlpnZ/wI451YDLwBrgLeA651zLd6Ywg3A28Ba4AXvWPAFmZvNrAjfGMSj3fHB2rqWquqaKK9t7I4iRURiRoersjrnrgySHPIL3Dl3B3BHkPQ3gTeDpG/GN5upWxX6DUpv2LOfwRnJ3f0WIiL9Vr+4QjqYwrxDwaFIg9IiIhHpt8Fh/JBDM5Y2alBaRCQi/TY4FAxKIzHeN1N2415d6yAiEol+Gxx8M5Z8XUtqOYiIRKbfBgeA8d64Q3ltI+U1Db1cGxGRo0e/Dg4T/MYdNCgtItJ5/To4+M9Y0jIaIiKd17+Dg9+1DkVagE9EpNP6dXAYPSidhLi2GUtqOYiIdFa/Dg5JCXGM8dZYUnAQEem8fh0c4NC4Q+n+BqrqtMaSiEhn9Pvg0O5KabUeREQ6pd8Hh8J2d4VTcBAR6Yx+Hxza3RVOM5ZERDql3weHgsFpxHszlnQhnIhI5/T74JCcEM/oQWmAFuATEemsfh8c4NAyGnuqG9hX39TLtRER6ftiIji0v/GPWg8iIh2JieAwXjOWREQiEhPBoVDXOoiIRCQmgsPY3HS8CUuazioi0gkxERxSEuMZPci3xpKms4qIdCwmggMcGnco2XeA/Qc0Y0lEJJyYCQ4T2s1YUutBRCScmAkO7QalNWNJRCSsmAkO7aaz6loHEZGwOgwOZvaYme01s1V+aQPNbJ6ZbfSec7x0M7P7zazIzFaY2RS/PHO84zea2Ry/9FPMbKWX534zs+7+kADjcjNoK1nTWUVEwutMy+EJYNZhabcA7zrnCoF3vdcAFwCF3mMu8CD4ggnwY+A0YBrw47aA4h0z1y/f4e/VLVKT4hk10FtjSd1KIiJhdRgcnHMfABWHJc8GnvS2nwQu8Ut/yvl8AmSb2TDgs8A851yFc64SmAfM8vYNcM79yznngKf8yup2bfd22FlVT01D85F6GxGRo160Yw55zrkSAO95iJc+Atjhd1yxlxYuvThI+hHhf1e4TepaEhEJqbsHpIONF7go0oMXbjbXzBaZ2aLS0tKIK9furnAKDiIiIUUbHPZ4XUJ4z3u99GJgpN9x+cCuDtLzg6QH5Zx7yDk31Tk3NTc3N+JK+98VbqOW0RARCSna4PAa0DbjaA7wql/61d6spdOBfV6309vA+WaW4w1Enw+87e3bb2ane7OUrvYrq9uNG5J+cFstBxGR0BI6OsDMngXOAQabWTG+WUd3Ai+Y2XXAduAy7/A3gQuBIqAOuAbAOVdhZrcDC73jfuacaxvk/ja+GVGpwF+9xxGRlpRAfk4qxZX1utZBRCSMDoODc+7KELtmBDnWAdeHKOcx4LEg6YuA4zuqR3eZkJdJcWU9xZX11DU2k5bU4Z9ARCTmxMwV0m3aBqWdg017a3u5NiIifVPMBQctoyEi0rGYCw6FebornIhIR2IuOLS/n7RaDiIiwcRccMhITmBEdiqgloOISCgxFxzgUOthe0UdB5paerk2IiJ9T0wGh7a7wjkHm0rVehAROVxMBgfdFU5EJLyYDA7j8zSdVUQknNgMDu1mLKnlICJyuJgMDgNSEhmWlQJoxpKISDAxGRzgUOthW3mtZiyJiBwmZoND26B0q4MtZVpjSUTEX+wGhzzdFU5EJJSYDQ4T8rSMhohIKDEbHMbn6loHEZFQYjY4ZKUlMiQzGdC1DiIih4vZ4AC+u8IBbC2vo6FZM5ZERNrEdHBom87a0urYWlbXy7UREek7Yjo4FGoZDRGRoGI7OGgBPhGRoGI8OKjlICISTEwHh5z0JAZneDOW1HIQETkopoMDHGo9bCmrpamltZdrIyLSN8R8cGi7Urq51bGtXGssiYiAggPj8w4NSm9Q15KICKDg0H5QWsFBRAToYnAws++b2WozW2Vmz5pZipmNMbP5ZrbRzJ43syTv2GTvdZG3v8CvnFu99PVm9tmufaTIaMaSiEigqIODmY0AvgdMdc4dD8QDVwC/BO5xzhUClcB1XpbrgErn3HjgHu84zGySl+84YBbwgJnFR1uvSA3KSGZQehKgloOISJuudislAKlmlgCkASXAecCL3v4ngUu87dnea7z9M8zMvPTnnHMNzrktQBEwrYv1ikjbMhqby2po1owlEZHog4Nzbifwa2A7vqCwD1gMVDnnmr3DioER3vYIYIeXt9k7fpB/epA87ZjZXDNbZGaLSktLo616gLZlNJpaHNsqtMaSiEhXupVy8P3qHwMMB9KBC4Ic6tqyhNgXKj0w0bmHnHNTnXNTc3NzI690CBPytIyGiIi/rnQrzQS2OOdKnXNNwEvAp4Bsr5sJIB/Y5W0XAyMBvP1ZQIV/epA8PWL8EN0VTkTEX1eCw3bgdDNL88YOZgBrgL8DX/KOmQO86m2/5r3G2/+ec8556Vd4s5nGAIXAgi7UK2LtFuDT/aRFREjo+JDgnHPzzexFYAnQDCwFHgLeAJ4zs597aY96WR4FnjazInwthiu8clab2Qv4AkszcL1zrkfvvDM4I4nstESq6poUHERE6EJwAHDO/Rj48WHJmwky28g5dwC4LEQ5dwB3dKUuXWFmTBiSyYKtFWwqraGl1REfF2woREQkNsT8FdJtxnszlhqbW9muGUsiEuMUHDyFGpQWETlIwcGjQWkRkUMUHDwT8tRyEBFpo+Dgyc1MZkCKb3xeLQcRiXUKDh4zo9C7Urpor2/GkohIrFJw8NPWtdTQ3EpxpWYsiUjsUnDwM36I1lgSEQEFh3ba3/hHwUFEYpeCg5/CPN0VTkQEFBzaGToghcxkb8aSupVEJIYpOPgxs4PLaBTtraFVM5ZEJEYpOBymbdyhvqmFnVX1vVwbEZHeoeBwmHZ3hdO4g4jEKAWHw4wZnH5w+4mPtlKkACEiMUjBwc+Wslp++trqg68/2FjGzN98wC1/XqErpkUkpig4eBqbW/n64wvYXhk4zvDcwh3c/+7GXqiViEjvUHDwvL16N9vKQy+Z8eTHWznQ1KN3LxUR6TUKDp7F2yrD7q+qb2JzaW0P1UZEpHcpOHiSEzv+U3TmGBGR/kDfdp7zJ+WF3T8uN52xfjOZRET6MwUHz5RROXwmTICYO30sZtaDNRIR6T0KDh4z47dXTubqM0aTnBD4Z/lgQ1kv1EpEpHeYc0fn/P2pU6e6RYsWHZGy99U3sWHPfuIMbn5+OdsqfLOYHr/mVM6dOOSIvKeISE8ws8XOuakdHaeWQxBZqYmcWjCQU0YP5PZLjj+Y/t+vrKK+UdNZRaT/U3DowPQJuXz+pOEAFFfWc58uhhORGNCl4GBm2Wb2opmtM7O1ZnaGmQ00s3lmttF7zvGONTO738yKzGyFmU3xK2eOd/xGM5vT1Q/V3f77omPJTPHd5+GRDzezfrfWWxKR/q2rLYf7gLecc8cAJwFrgVuAd51zhcC73muAC4BC7zEXeBDAzAYCPwZOA6YBP24LKH3FkMwUfjjrGACaWx3/+fJK3etBRPq1qIODmQ0ApgOPAjjnGp1zVcBs4EnvsCeBS7zt2cBTzucTINvMhgGfBeY55yqcc5XAPGBWtPU6Ur4ybRSTR2UDvqupn1+0o5drJCJy5HSl5TAWKAUeN7OlZvaImaUDec65EgDvuW16zwjA/xu12EsLlR7AzOaa2SIzW1RaWtqFqkcuLs74xaUnEB/nu9bhf95cS+n+hh6tg4hIT+lKcEgApgAPOucmA7Uc6kIKJtgVZC5MemCicw8556Y656bm5uZGWt8uO3bYAK47awwA1QeaueONNT1eBxGRntCV4FAMFDvn5nuvX8QXLPZ43UV4z3v9jh/plz8f2BUmvU+6aWYhI7JTAXhl2S7+uVEXx4lI/xN1cHDO7QZ2mNlEL2kGsAZ4DWibcTQHeNXbfg242pu1dDqwz+t2ehs438xyvIHo8720PiktKYGfzT7u4Ov/emWllvIWkX6nq7OVvgv80cxWACcDvwDuBD5jZhuBz3ivAd4ENgNFwMPAdwCccxXA7cBC7/EzL63PmnFsHrOOGwrA1vI6Hvh7US/XSESke2n5jCiV7Ktn5t3/oLaxhcR44683Tmf8kIxeq4+ISGdo+YwjbFhWKj/4rK9HranFcdvLKzlaA62IyOEUHLrg6jMKOGFEFgDzt1Tw5yU7e7lGIiLdQ8GhC+K9ax+8Sx+44401VNQ29m6lRES6gYJDF52Qn8XVZxQAUFnXxP+8ubZ3KyQi0g0UHLrBv58/gaEDUgD40+Ji5m8u7+UaiYh0jYJDN8hMSeQnF086+Po/X15JQ7OufRCRo1dCb1egv/jscUOZccwQ3l23l02ltVzz+EJGZKcyNCuFL0zJZ8zg9N6uoohIpyk4dBMz46ezj+ODjaU0tTg+3nSoa+l37xXx/2ZN5DvnjO/FGoqIdJ66lbpReU0jTS2B1zo44Fdvree9dXt6vlIiIlFQcOhGT368Nez+x/4Zfr+ISF+h4NCNVu3aF3b/yp3h94uI9BUKDt0oLSn8EE56UnwP1UREpGsUHLrR504YFnZ/RkoCjc2tPVQbEZHoKTh0oytPG8WEvNArs27YU8PXHp1PVZ2W2BCRvk3BoRtlJCfw3Nwz+NIp+SQl+P608XHG6WMHMiDF1+U0f0sFlz7wMVvLanuzqiIiYel+DkfI/gNNlO5vYFB6MllpiWwpq+WaxxewtbwOgJy0RB66eiqnFgzs5ZqKSCzR/Rx6WWZKImNzM8hKSwRgzOB0Xv7OmUzzgkFlXRNXPTyfV5ZqmW8R6XsUHHpQTnoST39jGpdOHgFAY0srNz2/jHv/tkE3ChKRPkXBoYclJ8Tzm8tP4vszJxxMu/dvG7n5heVarE9E+gwFh15gZtw4s5D7rjiZpHjfP8HLS3fytUcWUKmbBYlIH6Dg0ItmnzyCP37zNHK8cYkFWyu49IGP2FxaQ0urY/G2Ct5bt4fiyrperqmIxBrNVuoDtpbVcu0TC9nsTW9NT4onLSme0hpfK8KAmZPyuPMLJzAoI7kXayoiRzvNVjqKFAxO56XvfIrTx/pmMtU2thwMDOBb1XXemj1c/dgCXWEtIj1CwaGPyE5L4qlrT2NQelLIY1bvquat1bt7sFYiEqsUHPqQ+qYWyjsYkH5vre4JISJHnoJDH9Lc0nGXUVPr0TlGJCJHly4HBzOLN7OlZva693qMmc03s41m9ryZJXnpyd7rIm9/gV8Zt3rp683ss12t09FqYHoS44eEXrgPYEdFHfvqm3qoRiISq7qj5XAjsNbv9S+Be5xzhUAlcJ2Xfh1Q6ZwbD9zjHYeZTQKuAI4DZgEPmFlM3vjAzPj22ePCHrOieB+z7v2Aj4vKeqhWIhKLuhQczCwf+BzwiPfagPOAF71DngQu8bZne6/x9s/wjp8NPOeca3DObQGKgGldqdfR7Iun5PMfsyaSGG/t0vMykxmc4RusLtl3gK88Mp+f/WUNB5p0VbWIdL/wty7r2L3AfwCZ3utBQJVzrtl7XQyM8LZHADsAnHPNZrbPO34E8Ilfmf55YtJ3zhnP5VNH8taq3eyrb2LSsAFMn5BLzYFmfvTaKl5dtguAxz7awgcbS7n3yydz/IisXq61iPQnUbcczOwiYK9zbrF/cpBDXQf7wuU5/D3nmtkiM1tUWloaUX2PNoMzkvnq6aO5/tzxnHvMEOLjjKy0RO67YjK/vXIyWam+q6qL9tZwye8/4rfvbuzUgLaISGd0pVvpTOBiM9sKPIevO+leINvM2lok+cAub7sYGAng7c8CKvzTg+Rpxzn3kHNuqnNuam5ubheqfnT7/EnDefum6Xy6cDAAza2Ou+dt4LL/+xdbdBMhEekGUQcH59ytzrl851wBvgHl95xzVwF/B77kHTYHeNXbfs17jbf/Pedbu+M14ApvNtMYoBBYEG29YsXQrBSeunYat88+jpRE3z/j0u1VXHjfh/zhk23UNjTxx/nb+LenFzP3qUU88dEW9h/QLCcR6ZxuWVvJzM4BfuCcu8jMxuJrSQwElgJfdc41mFkK8DQwGV+L4Qrn3GYv/23AtUAzcJNz7q8dvWd/WlupqzaV1nDzC8tZvqPqYFpKYhwHmtp3Mw3LSuEP3ziNcbnhp8uKSP/V2bWVtPBeP9Hc0srv/76J+9/bSEuYC+WOGZrJX2/8NL6JYiISa7TwXoxJiI/jxpmF3H3ZSWGPW7d7Pwu2VPRQrUTkaKXg0M90pkGwrLiq44NEJKYpOPQzbVNcw7nrrfXc8MwSPtxYSqvWahKRILp6EZz0MZ8aN5jBGcmU1TSEPKa51fH6ihJeX1HCiOxULpuaz2VTRzIiOzXg2H31TRTtrSE9OZ6JeZkaqxCJERqQ7odeX7GL7z6zNOiVhLOOz2NdyX62lre/9agZnDV+MF8+dSSfmZRHayv8/I01vLi4mAbvBkPjctP5r89N4txjhvTApxCRI0GzlWLchxtL+e27RSzY6ht8PnlkNtefO57PTMrDOceCLRU8v2gHb64sCZjympOWSFpSAjur6gPKjTN44pppTJ8QuxchihzNFBwEgLrGZpyD9OTgPYjVB5r4y/JdvLBwB8uL93WqzBPzs3jthrO6s5oi0kM6Gxw05tDPpSWF/ycekJLIVaeN5qrTRrO2pJrnF+7g2QXbD3YlBbOieB8l++oZlhU4RiEi/YNmK8lBxw4bwE8uPo6LTxre4bHff34ZLy4u1o2HRPoptRwkwKkFA/nT4uKwx3yyuYJPNleQGG+cNX4wF54wjPMnDSUrrf1U2nW7q3lhYTE7q+oYlpXK5VNHMmn4gCNZfRHpBhpzkAD1jS2c++v32V19IOj+3MxkSvcHTpVNjDfOPBgo8vjj/O3c9fb6gOO+P3MCN84s7PZ6i0jHNCAtXbJ+936ufWJhwIylr50+mp9efBxby2v566rdvLGihDUl1QH5481oCXNuPf71UzUlVqQXKDhIlzU2t/LOmt2s3LmPjKQELjhhGOOHBK7ouqWsljdXlvDmyhJW7woMFMGcPSGXJ6+N2bvBivQaBQfpFVvLanlzVQn3zNtAU0vocyspPo5/O3ssUwsGMmV0Dhkhpto2t7Ty1mpfC6WmoZlJwwdw1bTRjBqUdqQ+gki/puAgverC+z4M2t0UTJzBpOEDOLVgIKcWDGRqQQ5DMlOoa2zm2icW8snm9qvIJsYbv71yCrOOH3okqi7Sr+k6B+lVF588PGxwMA7dKLzVwaqd1azaWc3jH20FoGBQGonxcWzcWxOQt6nFceNzS/nwh+cyJDOl+ysvIrrOQY6Mr54+mmOGZgbdNyEvg49vPY+nr5vG92YUcsbYQQdvddpma3ld0MDQpqG5lRc7mG4rItFTt5IcMVV1jdz9zgZeWlJMbWMLaUnxXDp5BD84fyI56Untjm1qaWX1rmoWbqlg4dYK5m+p6PACu5y0RL4wJZ8po3I4ZXQOQ7OCtyKWbq/kwfc38VFRGWa+6zK+fc44ThqZ3W2fVeRooTEH6TMam1vZV99EVmoiSQmda6xW1TZy8u3zInqfEdmpTB6VzSmjc5gyKodJwwfwj/Wl/NsfFtN82H0rEuKMh+dM5dyJmk4rsUXBQY561z2xkHfX7Q25PzHews6ISkmMo7nFBQSGNkMHpPDhD88lMb7jgOWcY9mOKtaW7CczJYFzjxkScoaVSF+mAWk56t164bEs3lZJVZDupavPGM1tnzuWNbuqWbytkqXbq1i8rbLdVd2HL0V+uN3VB3j4w81cOnkEQwekhLyRUXFlHTc8s5RlOw7dXjU9KZ5bLjiGr51REN2HE+nj1HKQPm1LWS33v7uRN1aW0NjcyrjcdK45cwxXnTYq6Jf5rqp6Fm+rZMn2Suat2UNxZeA9KYJJS4pnzOB0xuVmMDY3nbG5GYzLTWd4VipfePBjtpTVBs13/5WTO7VQoUhfoW4l6VdaWx2NLa2kJMZ3Os/Hm8r4ysPzj2CtfDOv3r5peoe3T21tdby5qoQXFhWze189I3PSDt51T7delZ6kbiXpV+LijJS4zgcGgNPHDGLs4HQ2h/jVPzwrhatOH8Xm0jo2l9WwaW8N1QeaI3qPDXtq+NbTi5k0fABjczMYOzidsbnp7e6j0dLquOn5Zfxl+a52+d5dt5crp43iF5cerwAhfY5aDtKvrdq5j68+Mj9g3GJgWhJ//OZpHDvs0PLhzjnKaxvZXFrL5tIaNpXW8JflJSFXpw1neFYK44b4gkVVXROv+gWGwz141RQuOGFYp8rdUlbLW6t2U9fYzIn52Zw7MZeETgyoi7RRt5KIZ2/1Af7wyTb+WVQGwKcLc7nq9FGdurr6laU7uen5ZSH3x5nvCu+uODE/i//96ikMHZBCXFzwFkRrq+Mnf1nNU//a1i59zOB0Hp0zlbG5gQsiigRzxIODmY0EngKGAq3AQ865+8xsIPA8UABsBS53zlWar918H3AhUAd83Tm3xCtrDvBfXtE/d8492dH7KzhIT2hobuGi+/8Z8mrtB6+azLQxg9hcVsumvTXtnrdX1NESQeRISohjZE4qBYPSGTUojdED0xg9OJ3RA9N4fUUJv5m3IWi+kQNT+dvNZ5Oc0Llut1U79/HJ5nLizDh7Yi7jFFhiSk8Eh2HAMOfcEjPLBBYDlwBfByqcc3ea2S1AjnPuh2Z2IfBdfMHhNOA+59xpXjBZBEzFt9zOYuAU51xluPdXcJCesqf6AN97dinztxxaADArNZHbLjyWy08dGTJfY3Mr2ytq+frjCzs9aypa91x+EpdOyQ97TPWBJr737FLeX1/aLv2Sk4dz5xdPjGiw3zlHXWMLyQlx6tY6yvR4t5KZvQr8znuc45wr8QLI+865iWb2f972s97x64Fz2h7OuW956e2OC0XBQXra2pJq1u2uJiM5kU8XDu70l+kjH27m52+sDbl/1nFDyUxJYFtFHdvL66Ia44g3Y/SgNEbkpJKfk0p+Thojsn3bI3JSGZKZwtynFoW8qPDKaaP4ny+c0OH7NLe08thHW3jqX9sorqwnOSGOz504jJs/M4H8HC2jfjTo0dlKZlYATAbmA3nOuRIAL0C0rU8wAtjhl63YSwuVLtKnHDtsQLsB7M762hmj+dvaPQFLjwPMOGYIv/vK5Ha/vg80tbC9oo5t5XVsK69l1c59vLIs9IA2QItzbC6rDTkzKyHOQl4pDvDCwu186+wxjB6YHnLmlHOOm19Yzmt+g+sNza28tGQnH2wo5eXvnMnIgZ0LEEu2V/LnxcXsqW6gYJBvWm9hXvCFGqV3dDk4mFkG8GfgJudcdZgpecF2uDDpwd5rLjAXYNSoUZFXVqQXJCfE88Q103jqX1v506Jidu87wIicVK6cNoqvnDYqoFsmJTGeCXmZTPC+LJ1zrCmpZsOe0KvUHjssk+r6Zkr21QcdIA8XGABaHJxz1z9ITohjcEYyA9OTGJSRxKD0ZO85ifKaxnaBwV9ZTSO/fmc9910xOez7OOe4/fW1PPbRlnbpj/5zCz/6/CSuOXNM2Pz+5SzZXsU7q3dzoKmFyaNymHX80Ii6xiS8LnUrmVki8DrwtnPuN17aetStJNKt5q3Zw9ynFgX91XTOxFwe//qpmBnNLa3srj5AcWU9Oyvrfc9VdSzeVsmm0uCtiu5iwOyTh/uCixdQBqYnMzA90XtO4r21e/j+C8tDlvHSdz7FlFE5Yd+nobmFG59dxlurd7dLH5GdyhPXnBpRC2RFcRU33T9GAAANpElEQVQLtlSQEGecM3EIBYPTO533aNUTA9IGPIlv8Pkmv/S7gHK/AemBzrn/MLPPATdwaED6fufcNG9AejEwxStiCb4B6cA2uB8FB4k1b60q4Y4317Kjwje4nZQQxxen5POjiyaRmhT+F/OOijqm/+rvwZvk+LqdZhw7hH31TZTXNFJR20hFXSM9PdP95JHZzJ0+lqzUxIOPAamJZCYnHJzme/vra3j0n1uC5s/PSeW9fz+nw9V/q+oauf6ZJXxUVN4u/fKp+dxx6QmdWowRfC2YRdsqKdpbQ05aEudMzO3zrZeeCA5nAR8CK/FNZQX4T3zjDi8Ao4DtwGXOuQovmPwOmIVvKus1zrlFXlnXenkB7nDOPd7R+ys4SCxqbXWs3lVNbWMzxwzNJDstqeNMnltfWsmzC7YH3XfTzEJumjmhXVpLq6OqrpHy2kbKahp4/KOtzFuzp0v1j1acQWZKIpkpCeysqg8btK4/dxyzjhtGdlpgYAHfF/pXHp7PvzaXB81/7Zlj+NHnJ3VYp6K9NdzwzBLW7d5/MC07NZGfzj6O2Sd3fth07/4DvLGihLKaBsYMzuDCE4a2u8K+u+kiOBFpp6mllTv/uo4/fLKNhmbf77mM5ATmTh/Ld88b3+ESHtvL6zjv7vdDjl/cOKOQq88YTUWtL6BUes8V3qO8tpG3V++msTn8arndrS2wZKUmkp2WiAHLi/eFPD4x3njp258iPyeNzJSEoFN199U1cf69/2BPdUPAPgOevu40zioc3GHdHv9oC794c227peezUxP5/VVTOHN8x/kB6htbeGXZTuat2UNjcyunjM7hqtNGMWRA8Is8FRxEJKiqukaWbq8iLs44ZXRORPeleGXpTn7wp+UBAWLmsXk8cNWUDrtz/ufNtfzfB5tD7v/qaaM4MT+bffVNQR9lNQ1H/JqRw6UnxR/s3hqQksiA1AT2VDewcmfoANN21Xt6cgIZyQnEB7ny/W9r9vCNp4J/h6UlxfP2TdM7nP1VVtPAVx7+JGCyQkZyAo9fcyqnFgwMyKPgICJHxJayWp5dsJ11u/eTnZrIxScN57xjhoRc+sNfRW0jlz7wEdvK6wL2nTI6h2e+eVrYK72dc8z8zT/CDq5/89NjSEmMDwwudb7nyrrGLi95EqnUxHgyUnyBIj05nozkBNaV7A96r5I2s47PY84ZY0hPjictKeHgc1pS/MExkblPLeKdEF19gzKS+OiH5wWMgSg4iEifVLq/gbvfWc+ry3ZR39RCVmoiXz51JDfNLOxUX/u7a/fwzacWBf2C/+KUfO6+/KSw+Tfu2c9n7vkg5P6keOPLp46itrGZ6vomquubqT7QRLUXZGobWzqs45GWFB9HSmJch6sI3/vlk7lkcvvxDy3ZLSJ9Um5mMnd+8UR+Nvt4ahqayUpNDNrtEsqMY/N4ZM5UfvXW+oODwdmpiVx9xmi+N6Oww/yFeZlcOnkELy/dGXT/v58/kW+dPS5k/uaWVh54v4jfzNsY8pjh2SmcM3EItQ3N1DY0s/9AM7WNzdQcaKamoYXahmbqm6IPMo0trTS2dDx2s6k09LUxHVFwEJFekZQQx8CEzs+28nfeMXmcO3EI28rrONDcQsGg9IimkN75xRPISE7guYXbDw4GZ6YkcP2545k7fWzYvAnxcXz9zDE8t2AHu/YFLnUSZ/Dry07iU+PCDyjf8MwSXl9REnL/RScO46T8bGobm6lr9AUU/+fymgbW+s2UCiYngtlsh1O3kojErPKaBpYXV5EQF8fUgpyIppBuKavle88ubTcwPTgjiZ9efDyfO7Hj+3Ns3LOf2b//iLog3VRjc9P5yw1nkR5msoBzjgvu+7DdVFp/CXHGx7ecFzBrSWMOIiJHmHOO5cX7KNpbw8D0RM4an9vhjC1/y3dU8d+vrmKFN7XW8HWb/eLS40NORfW3aGsFX310PgeaAruY/vPCY5g7PbB7TMFBROQoUbR3P2U1jYwelMawrNSI8q7auY/fvreRd9fupbnVcVJ+Ft86exwXhri7oIKDiEgMaW11tDjX4dIfmq0kIhJD4uKMuKCLXEdZXreVJCIi/YaCg4iIBFBwEBGRAAoOIiISQMFBREQCKDiIiEgABQcREQlw1F4EZ2alwLYwhwwGyrr4Nn2hjL5Qh75SRl+oQ3eU0Rfq0FfK6At16Ctl9FQdRjvncjssyTnXLx/Aov5QRl+oQ18poy/UQZ9Df4v+/rdoe6hbSUREAig4iIhIgP4cHB7qJ2X0hTr0lTL6Qh26o4y+UIe+UkZfqENfKaMv1OGgo3ZAWkREjpz+3HIQEZEo9bvgYGaPmdleM1sVZf6RZvZ3M1trZqvN7MYoykgxswVmttwr46fR1MUrK97MlprZ61Hm32pmK81smZlFfAMMM8s2sxfNbJ33NzkjwvwTvfdue1Sb2U1R1OP73t9ylZk9a2Yd3yarff4bvbyrI3n/YOeTmQ00s3lmttF7zokw/2VePVrNrOObrgQv4y7v32SFmb1sZtlRlHG7l3+Zmb1jZsMjLcNv3w/MzJlZyBsnh6jDT8xsp9/5cWE0dTCz75rZeu/v+qtIyzCz5/3qsNXMlkWY/2Qz+6Tt/5mZTYuiDieZ2b+8/69/MbMBHZQR9LsqkvMzrO6a9tRXHsB0YAqwKsr8w4Ap3nYmsAGYFGEZBmR424nAfOD0KOtzM/AM8HqU+bcCg7vw93wS+Ia3nQRkd6GseGA3vnnWkeQbAWwBUr3XLwBfjyD/8cAqIA3fPUz+BhRGez4BvwJu8bZvAX4ZYf5jgYnA+8DUKOtwPpDgbf8yXB3ClDHAb/t7wP9GWoaXPhJ4G991RyHPtRB1+Anwgwj+LYOVca73b5rsvR4Szefw23838KMI6/AOcIG3fSHwfhSfYyFwtrd9LXB7B2UE/a6K5PwM9+h3LQfn3AdARRfylzjnlnjb+4G1+L6cIinDOedqvJeJ3iPiwR0zywc+BzwSad7u4P1ymQ48CuCca3TOVXWhyBnAJudcuIsXQ0kAUs0sAd+X/K4I8h4LfOKcq3PONQP/AC7tTMYQ59NsfEET7/mSSPI759Y659Z3su6hynjH+ywAnwD5UZRR7fcynQ7O0TD/t+4B/qML+TstRBnfBu50zjV4x+yNth5mZsDlwLMR5ndA2y/9LDo4P0OUMRH4wNueB3yxgzJCfVd1+vwMp98Fh+5kZgXAZHy//CPNG+81TfcC85xzEZcB3IvvP13g3cM7zwHvmNliM5sbYd6xQCnwuNe19YiZpXehLlcQ5j9dKM65ncCvge1ACbDPOfdOBEWsAqab2SAzS8P3y25kpPXwk+ecK/HqVgIM6UJZ3eFa4K/RZDSzO8xsB3AV8KMo8l8M7HTOLY/m/T03eN1bj0XZBTIB+LSZzTezf5jZqV2oy6eBPc65jRHmuwm4y/tb/hq4NYr3XgVc7G1fRgTn6GHfVd1yfio4hGBmGcCfgZsO+4XVKc65Fufcyfh+0U0zs+MjfP+LgL3OucWRvvdhznTOTQEuAK43s+kR5E3A1/R90Dk3GajF10yNmJkl4Tvx/xRF3hx8v4bGAMOBdDP7amfzO+fW4ut6mQe8BSwHmsNmOkqY2W34Pssfo8nvnLvNOTfSy39DhO+dBtxGFEHFz4PAOOBkfIH/7ijKSABygNOB/we84LUAonElUfyAwdd6+b73t/w+Xms7Qtfi+z+6GF83UWNnMnX1uyoUBYcgzCwR3x/7j865l7pSltcN8z4wK8KsZwIXm9lW4DngPDP7QxTvv8t73gu8DIQdKDtMMVDs1+p5EV+wiMYFwBLn3J4o8s4EtjjnSp1zTcBLwKciKcA596hzbopzbjq+5nykvwz97TGzYQDec9hujCPFzOYAFwFXOa+DuQueoYNujCDG4QvYy73zNB9YYmZDO1uAc26P90OqFXiYyM7PNsXAS1537gJ8Le2QA+OheF2WXwCej6IOc/Cdl+D7ARTx53DOrXPOne+cOwVfgNrUUZ4Q31Xdcn4qOBzG+8XxKLDWOfebKMvIbZs9Ymap+L7c1kVShnPuVudcvnOuAF93zHvOuU7/WvbeO93MMtu28Q1idnoWl3NuN7DDzCZ6STOANZHUwU+0v8jA1510upmlef8+M/D1r3aamQ3xnkfh+wKIti4Ar+H7MsB7frULZUXFzGYBPwQuds7VRVlGod/Li4n8HF3pnBvinCvwztNifAOkuyOowzC/l5cSwfnp5xXgPK+8CfgmTkSzgN1MYJ1zrjiKvLuAs73t84jix4ffORoH/Bfwvx0cH+q7qnvOz2hGsfvyA99/+hKgCd/Jel2E+c/C10+/AljmPS6MsIwTgaVeGasIM/Ohk+WdQxSzlfCNGSz3HquB26Io42RgkfdZXgFyoigjDSgHsrrwN/gpvi+vVcDTeDNTIsj/Ib7AthyY0ZXzCRgEvIvvC+BdYGCE+S/1thuAPcDbUdShCNjhd452NNMoWBl/9v6eK4C/ACMiLeOw/VsJP1spWB2eBlZ6dXgNGBbF50gC/uB9liXAedF8DuAJ4N+iPCfOAhZ759d84JQoyrgR34yjDcCdeBcphykj6HdVJOdnuIeukBYRkQDqVhIRkQAKDiIiEkDBQUREAig4iIhIAAUHEREJoOAgIiIBFBxERCSAgoOIiAT4//rRzOGuP7LJAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pointplot(x = list(wcss.keys()), y = list(wcss.values()))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KMeans(algorithm='auto', copy_x=True, init='k-means++', max_iter=300,\n",
       "    n_clusters=4, n_init=10, n_jobs=1, precompute_distances='auto',\n",
       "    random_state=None, tol=0.0001, verbose=0)"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# choose n_clusters = 4\n",
    "clus = KMeans(n_clusters= 4, init= 'k-means++', max_iter= 300)\n",
    "clus.fit(rfm_scaled)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "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>MonetaryValue</th>\n",
       "      <th>R</th>\n",
       "      <th>F</th>\n",
       "      <th>M</th>\n",
       "      <th>RFM_Segment</th>\n",
       "      <th>RFM_Score</th>\n",
       "      <th>RFM_Level</th>\n",
       "      <th>Cluster</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CustomerID</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",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12347.0</th>\n",
       "      <td>2</td>\n",
       "      <td>29</td>\n",
       "      <td>641.00</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>444</td>\n",
       "      <td>12.0</td>\n",
       "      <td>Top</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12348.0</th>\n",
       "      <td>248</td>\n",
       "      <td>3</td>\n",
       "      <td>400.00</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>114</td>\n",
       "      <td>6.0</td>\n",
       "      <td>Mid</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12349.0</th>\n",
       "      <td>18</td>\n",
       "      <td>16</td>\n",
       "      <td>324.42</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>433</td>\n",
       "      <td>10.0</td>\n",
       "      <td>Top</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12350.0</th>\n",
       "      <td>310</td>\n",
       "      <td>5</td>\n",
       "      <td>110.40</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>122</td>\n",
       "      <td>5.0</td>\n",
       "      <td>Low</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12352.0</th>\n",
       "      <td>36</td>\n",
       "      <td>17</td>\n",
       "      <td>621.28</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>334</td>\n",
       "      <td>10.0</td>\n",
       "      <td>Top</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Recency  Frequency  MonetaryValue  R  F  M RFM_Segment  RFM_Score  \\\n",
       "CustomerID                                                                      \n",
       "12347.0           2         29         641.00  4  4  4         444       12.0   \n",
       "12348.0         248          3         400.00  1  1  4         114        6.0   \n",
       "12349.0          18         16         324.42  4  3  3         433       10.0   \n",
       "12350.0         310          5         110.40  1  2  2         122        5.0   \n",
       "12352.0          36         17         621.28  3  3  4         334       10.0   \n",
       "\n",
       "           RFM_Level  Cluster  \n",
       "CustomerID                     \n",
       "12347.0          Top        2  \n",
       "12348.0          Mid        1  \n",
       "12349.0          Top        1  \n",
       "12350.0          Low        0  \n",
       "12352.0          Top        1  "
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Assign the clusters to datamart\n",
    "datamart['Cluster'] = clus.labels_\n",
    "datamart.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3-3. Visualization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can go further analysis with this label as we did with RFM quantils."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>Recency</th>\n",
       "      <th>Frequency</th>\n",
       "      <th colspan=\"2\" halign=\"left\">MonetaryValue</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>mean</th>\n",
       "      <th>mean</th>\n",
       "      <th>mean</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cluster</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>182.782</td>\n",
       "      <td>3.186</td>\n",
       "      <td>52.478</td>\n",
       "      <td>1236</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>106.371</td>\n",
       "      <td>15.879</td>\n",
       "      <td>315.695</td>\n",
       "      <td>1288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>13.094</td>\n",
       "      <td>59.665</td>\n",
       "      <td>1570.621</td>\n",
       "      <td>842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>18.625</td>\n",
       "      <td>8.925</td>\n",
       "      <td>116.571</td>\n",
       "      <td>755</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Recency Frequency MonetaryValue      \n",
       "            mean      mean          mean count\n",
       "Cluster                                       \n",
       "0        182.782     3.186        52.478  1236\n",
       "1        106.371    15.879       315.695  1288\n",
       "2         13.094    59.665      1570.621   842\n",
       "3         18.625     8.925       116.571   755"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "datamart.groupby('Cluster').agg({\n",
    "    'Recency' : 'mean',\n",
    "    'Frequency' : 'mean',\n",
    "    'MonetaryValue' : ['mean', 'count']\n",
    "}).round(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In marketing, snail plot and heatmap are often used plot for visualization. I'll use the `rfm_scaled` dataframe with normalized rfm values for the plot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "# assign cluster column \n",
    "rfm_scaled['Cluster'] = clus.labels_\n",
    "rfm_scaled.reset_index(inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "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>CustomerID</th>\n",
       "      <th>Recency</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>MonetaryValue</th>\n",
       "      <th>Cluster</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>12347.0</td>\n",
       "      <td>-2.104896</td>\n",
       "      <td>0.951625</td>\n",
       "      <td>1.085161</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>12348.0</td>\n",
       "      <td>1.189415</td>\n",
       "      <td>-0.920178</td>\n",
       "      <td>0.752902</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>12349.0</td>\n",
       "      <td>-0.603319</td>\n",
       "      <td>0.461391</td>\n",
       "      <td>0.605778</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>12350.0</td>\n",
       "      <td>1.342511</td>\n",
       "      <td>-0.499270</td>\n",
       "      <td>-0.153068</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>12352.0</td>\n",
       "      <td>-0.128993</td>\n",
       "      <td>0.510909</td>\n",
       "      <td>1.063339</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerID   Recency  Frequency  MonetaryValue  Cluster\n",
       "0     12347.0 -2.104896   0.951625       1.085161        2\n",
       "1     12348.0  1.189415  -0.920178       0.752902        1\n",
       "2     12349.0 -0.603319   0.461391       0.605778        1\n",
       "3     12350.0  1.342511  -0.499270      -0.153068        0\n",
       "4     12352.0 -0.128993   0.510909       1.063339        1"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfm_scaled.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "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>CustomerID</th>\n",
       "      <th>Cluster</th>\n",
       "      <th>Metrics</th>\n",
       "      <th>Value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>12347.0</td>\n",
       "      <td>2</td>\n",
       "      <td>Recency</td>\n",
       "      <td>-2.104896</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>12348.0</td>\n",
       "      <td>1</td>\n",
       "      <td>Recency</td>\n",
       "      <td>1.189415</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>12349.0</td>\n",
       "      <td>1</td>\n",
       "      <td>Recency</td>\n",
       "      <td>-0.603319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>12350.0</td>\n",
       "      <td>0</td>\n",
       "      <td>Recency</td>\n",
       "      <td>1.342511</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>12352.0</td>\n",
       "      <td>1</td>\n",
       "      <td>Recency</td>\n",
       "      <td>-0.128993</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerID  Cluster  Metrics     Value\n",
       "0     12347.0        2  Recency -2.104896\n",
       "1     12348.0        1  Recency  1.189415\n",
       "2     12349.0        1  Recency -0.603319\n",
       "3     12350.0        0  Recency  1.342511\n",
       "4     12352.0        1  Recency -0.128993"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# melt the dataframe\n",
    "rfm_melted = pd.melt(frame= rfm_scaled, id_vars= ['CustomerID', 'Cluster'], var_name = 'Metrics', value_name = 'Value')\n",
    "rfm_melted.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\jjone\\Anaconda3\\lib\\site-packages\\scipy\\stats\\stats.py:1713: FutureWarning: Using a non-tuple sequence for multidimensional indexing is deprecated; use `arr[tuple(seq)]` instead of `arr[seq]`. In the future this will be interpreted as an array index, `arr[np.array(seq)]`, which will result either in an error or a different result.\n",
      "  return np.add.reduce(sorted[indexer] * weights, axis=axis) / sumval\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x2dbeb04ad68>"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# a snake plot\n",
    "sns.lineplot(x = 'Metrics', y = 'Value', hue = 'Cluster', data = rfm_melted)\n",
    "plt.title('Snake Plot of RFM')\n",
    "plt.legend(loc = 'upper right')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Heatmap is efficient for comparing the standardized values. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "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>MonetaryValue</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cluster</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>182.782362</td>\n",
       "      <td>3.186084</td>\n",
       "      <td>52.477872</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>106.371118</td>\n",
       "      <td>15.878882</td>\n",
       "      <td>315.694961</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>13.093824</td>\n",
       "      <td>59.665083</td>\n",
       "      <td>1570.620784</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>18.625166</td>\n",
       "      <td>8.924503</td>\n",
       "      <td>116.571457</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Recency  Frequency  MonetaryValue\n",
       "Cluster                                      \n",
       "0        182.782362   3.186084      52.477872\n",
       "1        106.371118  15.878882     315.694961\n",
       "2         13.093824  59.665083    1570.620784\n",
       "3         18.625166   8.924503     116.571457"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# the mean value for each cluster\n",
    "cluster_avg = datamart.groupby('Cluster').mean().iloc[:, 0:3]\n",
    "cluster_avg.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Recency           94.154817\n",
       "Frequency         19.744237\n",
       "MonetaryValue    456.673601\n",
       "dtype: float64"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# the mean value in total \n",
    "total_avg = datamart.iloc[:, 0:3].mean()\n",
    "total_avg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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>MonetaryValue</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cluster</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.941296</td>\n",
       "      <td>-0.838632</td>\n",
       "      <td>-0.885087</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.129747</td>\n",
       "      <td>-0.195771</td>\n",
       "      <td>-0.308708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.860933</td>\n",
       "      <td>2.021899</td>\n",
       "      <td>2.439263</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.802186</td>\n",
       "      <td>-0.547995</td>\n",
       "      <td>-0.744738</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Recency  Frequency  MonetaryValue\n",
       "Cluster                                    \n",
       "0        0.941296  -0.838632      -0.885087\n",
       "1        0.129747  -0.195771      -0.308708\n",
       "2       -0.860933   2.021899       2.439263\n",
       "3       -0.802186  -0.547995      -0.744738"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# the proportional mean value\n",
    "prop_rfm = cluster_avg/total_avg - 1\n",
    "prop_rfm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# heatmap\n",
    "sns.heatmap(prop_rfm, cmap= 'Blues', fmt= '.2f', annot = True)\n",
    "plt.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}