CustomersSegmentation.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "CustomersSegmentation.ipynb",
      "provenance": [],
      "toc_visible": true,
      "authorship_tag": "ABX9TyOBDLsfJWcTHhGytY40807J",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/zahadev/data_analysis/blob/master/CustomersSegmentation.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wNM3RGACWyjg"
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns; sns.set(style='white')\n",
        "%matplotlib inline\n",
        "from sklearn import decomposition\n",
        "from sklearn.cluster import KMeans\n",
        "from sklearn import metrics"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "r_V8-WTS89kg"
      },
      "source": [
        "## Кластеризация"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "587vTW1oXLTJ"
      },
      "source": [
        "bank = pd.read_csv(\"BankChurners.csv\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "J0NQ7_bms4Pt"
      },
      "source": [
        "Мы разделим клиентов на группы на основании возраста, пола, образование, семейнего статуса, дохода, периода отношение с банком, кредитнего лимита, количеством транзакции и ушел ли клиент.\n",
        "\n",
        "Что бы делать рекомендации более персонализированее."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "sYPq-gt1Xflx",
        "outputId": "4bc8f86b-ae1e-439f-fd44-2d53e1490135"
      },
      "source": [
        "# оставим только нужные нам данные и переименуем колонок\n",
        "bank = bank[['Attrition_Flag', 'Customer_Age', 'Gender', 'Education_Level', 'Marital_Status', 'Income_Category', 'Months_on_book', 'Credit_Limit', 'Total_Trans_Ct']]\n",
        "bank.columns = ['att_flag', 'age', 'gender', 'edu_level', 'marital_status', 'income', 'months_on_book', 'credit_limit', 'total_trans_ct']\n",
        "bank.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>att_flag</th>\n",
              "      <th>age</th>\n",
              "      <th>gender</th>\n",
              "      <th>edu_level</th>\n",
              "      <th>marital_status</th>\n",
              "      <th>income</th>\n",
              "      <th>months_on_book</th>\n",
              "      <th>credit_limit</th>\n",
              "      <th>total_trans_ct</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>45</td>\n",
              "      <td>M</td>\n",
              "      <td>High School</td>\n",
              "      <td>Married</td>\n",
              "      <td>$60K - $80K</td>\n",
              "      <td>39</td>\n",
              "      <td>12691.0</td>\n",
              "      <td>42</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>49</td>\n",
              "      <td>F</td>\n",
              "      <td>Graduate</td>\n",
              "      <td>Single</td>\n",
              "      <td>Less than $40K</td>\n",
              "      <td>44</td>\n",
              "      <td>8256.0</td>\n",
              "      <td>33</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>51</td>\n",
              "      <td>M</td>\n",
              "      <td>Graduate</td>\n",
              "      <td>Married</td>\n",
              "      <td>$80K - $120K</td>\n",
              "      <td>36</td>\n",
              "      <td>3418.0</td>\n",
              "      <td>20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>40</td>\n",
              "      <td>F</td>\n",
              "      <td>High School</td>\n",
              "      <td>Unknown</td>\n",
              "      <td>Less than $40K</td>\n",
              "      <td>34</td>\n",
              "      <td>3313.0</td>\n",
              "      <td>20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>40</td>\n",
              "      <td>M</td>\n",
              "      <td>Uneducated</td>\n",
              "      <td>Married</td>\n",
              "      <td>$60K - $80K</td>\n",
              "      <td>21</td>\n",
              "      <td>4716.0</td>\n",
              "      <td>28</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "            att_flag  age gender  ... months_on_book credit_limit total_trans_ct\n",
              "0  Existing Customer   45      M  ...             39      12691.0             42\n",
              "1  Existing Customer   49      F  ...             44       8256.0             33\n",
              "2  Existing Customer   51      M  ...             36       3418.0             20\n",
              "3  Existing Customer   40      F  ...             34       3313.0             20\n",
              "4  Existing Customer   40      M  ...             21       4716.0             28\n",
              "\n",
              "[5 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 218
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 241
        },
        "id": "Iyz-G7Q_mgcd",
        "outputId": "c380167d-3048-4b2a-f383-81db9fe27df7"
      },
      "source": [
        "#переобразуем категоральные данные на числовые\n",
        "bank = pd.get_dummies(bank, drop_first = True)\n",
        "bank.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>age</th>\n",
              "      <th>months_on_book</th>\n",
              "      <th>credit_limit</th>\n",
              "      <th>total_trans_ct</th>\n",
              "      <th>att_flag_Existing Customer</th>\n",
              "      <th>gender_M</th>\n",
              "      <th>edu_level_Doctorate</th>\n",
              "      <th>edu_level_Graduate</th>\n",
              "      <th>edu_level_High School</th>\n",
              "      <th>edu_level_Post-Graduate</th>\n",
              "      <th>edu_level_Uneducated</th>\n",
              "      <th>edu_level_Unknown</th>\n",
              "      <th>marital_status_Married</th>\n",
              "      <th>marital_status_Single</th>\n",
              "      <th>marital_status_Unknown</th>\n",
              "      <th>income_$40K - $60K</th>\n",
              "      <th>income_$60K - $80K</th>\n",
              "      <th>income_$80K - $120K</th>\n",
              "      <th>income_Less than $40K</th>\n",
              "      <th>income_Unknown</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>45</td>\n",
              "      <td>39</td>\n",
              "      <td>12691.0</td>\n",
              "      <td>42</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>49</td>\n",
              "      <td>44</td>\n",
              "      <td>8256.0</td>\n",
              "      <td>33</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>51</td>\n",
              "      <td>36</td>\n",
              "      <td>3418.0</td>\n",
              "      <td>20</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>40</td>\n",
              "      <td>34</td>\n",
              "      <td>3313.0</td>\n",
              "      <td>20</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>40</td>\n",
              "      <td>21</td>\n",
              "      <td>4716.0</td>\n",
              "      <td>28</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   age  months_on_book  ...  income_Less than $40K  income_Unknown\n",
              "0   45              39  ...                      0               0\n",
              "1   49              44  ...                      1               0\n",
              "2   51              36  ...                      0               0\n",
              "3   40              34  ...                      1               0\n",
              "4   40              21  ...                      0               0\n",
              "\n",
              "[5 rows x 20 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 219
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RxcGd93tu8fa"
      },
      "source": [
        "Так как у нас есть разные по масштабу переменные, нужно их нормировать"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P7STe-5NpWkC"
      },
      "source": [
        "# нормируем данные\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "scaler = StandardScaler()\n",
        "scaled_bank = scaler.fit_transform(bank)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lhkuZSRhvabJ"
      },
      "source": [
        "После нескольких экспериментов, и подсчетов метрик выбрал иерахическая кластеризацию, точнее алгомеративный метод кластеризации. \n",
        "\n",
        "При построение кластера наилушим способом определение расстояние между кластерами оказалось *Averge Linkage*"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 280
        },
        "id": "BjCKaywasD9_",
        "outputId": "3efb883f-68c6-4fd0-81bd-da3d63096d9b"
      },
      "source": [
        "mergings = linkage(scaled_bank, method='average')\n",
        "dendrogram(mergings,\n",
        "             leaf_rotation = 90,\n",
        "             leaf_font_size = 7,\n",
        "             )\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2KlWNvlQtvL7",
        "outputId": "1e86c5d3-2f4b-4164-cd45-6c67453779c8"
      },
      "source": [
        "from sklearn.cluster import AgglomerativeClustering\n",
        "amodel = AgglomerativeClustering(linkage='average', n_clusters =21)\n",
        "amodel = amodel.fit(scaled_bank)\n",
        "apred = amodel.labels_\n",
        "apred"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([4, 1, 5, ..., 1, 3, 1])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 256
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 768
        },
        "id": "cN9pOYuXukiX",
        "outputId": "2823933a-358b-44f2-b91d-9c202d016cae"
      },
      "source": [
        "res = bank.copy()\n",
        "res[\"cluster\"] = amodel.labels_\n",
        "res.groupby(\"cluster\").mean()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>age</th>\n",
              "      <th>months_on_book</th>\n",
              "      <th>credit_limit</th>\n",
              "      <th>total_trans_ct</th>\n",
              "      <th>att_flag_Existing Customer</th>\n",
              "      <th>gender_M</th>\n",
              "      <th>edu_level_Doctorate</th>\n",
              "      <th>edu_level_Graduate</th>\n",
              "      <th>edu_level_High School</th>\n",
              "      <th>edu_level_Post-Graduate</th>\n",
              "      <th>edu_level_Uneducated</th>\n",
              "      <th>edu_level_Unknown</th>\n",
              "      <th>marital_status_Married</th>\n",
              "      <th>marital_status_Single</th>\n",
              "      <th>marital_status_Unknown</th>\n",
              "      <th>income_$40K - $60K</th>\n",
              "      <th>income_$60K - $80K</th>\n",
              "      <th>income_$80K - $120K</th>\n",
              "      <th>income_Less than $40K</th>\n",
              "      <th>income_Unknown</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cluster</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>0</th>\n",
              "      <td>47.460993</td>\n",
              "      <td>36.947991</td>\n",
              "      <td>8334.625532</td>\n",
              "      <td>63.650118</td>\n",
              "      <td>0.789598</td>\n",
              "      <td>0.425532</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.484634</td>\n",
              "      <td>0.430260</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.158392</td>\n",
              "      <td>0.125296</td>\n",
              "      <td>0.130024</td>\n",
              "      <td>0.349882</td>\n",
              "      <td>0.153664</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>46.556967</td>\n",
              "      <td>36.145296</td>\n",
              "      <td>3779.779992</td>\n",
              "      <td>66.253275</td>\n",
              "      <td>0.836443</td>\n",
              "      <td>0.081778</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.415244</td>\n",
              "      <td>0.247320</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.211592</td>\n",
              "      <td>0.492656</td>\n",
              "      <td>0.434299</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.990075</td>\n",
              "      <td>0.001191</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>45.498938</td>\n",
              "      <td>35.318471</td>\n",
              "      <td>8593.454989</td>\n",
              "      <td>65.195329</td>\n",
              "      <td>0.830149</td>\n",
              "      <td>0.481953</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.511677</td>\n",
              "      <td>0.401274</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.222930</td>\n",
              "      <td>0.157113</td>\n",
              "      <td>0.140127</td>\n",
              "      <td>0.335456</td>\n",
              "      <td>0.087049</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>46.308511</td>\n",
              "      <td>35.821809</td>\n",
              "      <td>5707.666489</td>\n",
              "      <td>69.172872</td>\n",
              "      <td>0.821809</td>\n",
              "      <td>0.265957</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.345745</td>\n",
              "      <td>0.196809</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.223404</td>\n",
              "      <td>0.154255</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.321809</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.047872</td>\n",
              "      <td>0.595745</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>45.957976</td>\n",
              "      <td>35.536878</td>\n",
              "      <td>10596.796055</td>\n",
              "      <td>63.438250</td>\n",
              "      <td>0.873070</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.331904</td>\n",
              "      <td>0.244425</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.155232</td>\n",
              "      <td>0.163808</td>\n",
              "      <td>0.518010</td>\n",
              "      <td>0.402230</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>47.062500</td>\n",
              "      <td>36.661990</td>\n",
              "      <td>16877.929911</td>\n",
              "      <td>62.220663</td>\n",
              "      <td>0.839923</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.412628</td>\n",
              "      <td>0.263393</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.179209</td>\n",
              "      <td>0.541454</td>\n",
              "      <td>0.389668</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.002551</td>\n",
              "      <td>0.687500</td>\n",
              "      <td>0.007015</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>45.534884</td>\n",
              "      <td>34.837209</td>\n",
              "      <td>11853.293023</td>\n",
              "      <td>64.418605</td>\n",
              "      <td>0.720930</td>\n",
              "      <td>0.558140</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.139535</td>\n",
              "      <td>0.069767</td>\n",
              "      <td>0.348837</td>\n",
              "      <td>0.255814</td>\n",
              "      <td>0.116279</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>46.301087</td>\n",
              "      <td>35.780435</td>\n",
              "      <td>9409.070652</td>\n",
              "      <td>66.214130</td>\n",
              "      <td>0.833696</td>\n",
              "      <td>0.043478</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.333696</td>\n",
              "      <td>0.223913</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.189130</td>\n",
              "      <td>0.150000</td>\n",
              "      <td>0.471739</td>\n",
              "      <td>0.438043</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>44.538462</td>\n",
              "      <td>34.515385</td>\n",
              "      <td>18012.863846</td>\n",
              "      <td>63.646154</td>\n",
              "      <td>0.884615</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.300000</td>\n",
              "      <td>0.300000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.007692</td>\n",
              "      <td>0.253846</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.776923</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>44.250000</td>\n",
              "      <td>34.892857</td>\n",
              "      <td>9601.185714</td>\n",
              "      <td>65.821429</td>\n",
              "      <td>0.785714</td>\n",
              "      <td>0.500000</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.107143</td>\n",
              "      <td>0.214286</td>\n",
              "      <td>0.071429</td>\n",
              "      <td>0.357143</td>\n",
              "      <td>0.178571</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>44.704225</td>\n",
              "      <td>33.760563</td>\n",
              "      <td>9650.683099</td>\n",
              "      <td>70.760563</td>\n",
              "      <td>0.802817</td>\n",
              "      <td>0.014085</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.309859</td>\n",
              "      <td>0.267606</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.140845</td>\n",
              "      <td>0.098592</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>32.000000</td>\n",
              "      <td>36.000000</td>\n",
              "      <td>33711.000000</td>\n",
              "      <td>134.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>32.400000</td>\n",
              "      <td>19.000000</td>\n",
              "      <td>14653.400000</td>\n",
              "      <td>54.500000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.900000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.500000</td>\n",
              "      <td>0.100000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.400000</td>\n",
              "      <td>0.500000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.200000</td>\n",
              "      <td>0.300000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>26.000000</td>\n",
              "      <td>13.000000</td>\n",
              "      <td>34516.000000</td>\n",
              "      <td>59.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>46.505408</td>\n",
              "      <td>36.050147</td>\n",
              "      <td>8402.973451</td>\n",
              "      <td>65.024582</td>\n",
              "      <td>0.837758</td>\n",
              "      <td>0.437561</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.480826</td>\n",
              "      <td>0.425762</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.217306</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.194690</td>\n",
              "      <td>0.481809</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>61.000000</td>\n",
              "      <td>54.500000</td>\n",
              "      <td>7936.150000</td>\n",
              "      <td>31.500000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.500000</td>\n",
              "      <td>0.500000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>28.000000</td>\n",
              "      <td>14.571429</td>\n",
              "      <td>2538.614286</td>\n",
              "      <td>34.714286</td>\n",
              "      <td>0.428571</td>\n",
              "      <td>0.714286</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.428571</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.428571</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>62.000000</td>\n",
              "      <td>56.000000</td>\n",
              "      <td>3477.000000</td>\n",
              "      <td>40.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>28.833333</td>\n",
              "      <td>18.500000</td>\n",
              "      <td>8014.666667</td>\n",
              "      <td>90.833333</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.166667</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.500000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.166667</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.833333</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>45.746468</td>\n",
              "      <td>35.507064</td>\n",
              "      <td>5502.694976</td>\n",
              "      <td>64.713501</td>\n",
              "      <td>0.854003</td>\n",
              "      <td>0.432496</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.399529</td>\n",
              "      <td>0.257457</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.001570</td>\n",
              "      <td>0.205651</td>\n",
              "      <td>0.485086</td>\n",
              "      <td>0.430926</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.987441</td>\n",
              "      <td>0.001570</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.004710</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>20</th>\n",
              "      <td>46.204301</td>\n",
              "      <td>36.010753</td>\n",
              "      <td>11150.084946</td>\n",
              "      <td>61.602151</td>\n",
              "      <td>0.849462</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.376344</td>\n",
              "      <td>0.236559</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.139785</td>\n",
              "      <td>0.139785</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "               age  months_on_book  ...  income_Less than $40K  income_Unknown\n",
              "cluster                             ...                                       \n",
              "0        47.460993       36.947991  ...               0.349882        0.153664\n",
              "1        46.556967       36.145296  ...               0.990075        0.001191\n",
              "2        45.498938       35.318471  ...               0.335456        0.087049\n",
              "3        46.308511       35.821809  ...               0.595745        0.000000\n",
              "4        45.957976       35.536878  ...               0.000000        0.000000\n",
              "5        47.062500       36.661990  ...               0.007015        0.000000\n",
              "6        45.534884       34.837209  ...               0.255814        0.116279\n",
              "7        46.301087       35.780435  ...               0.000000        1.000000\n",
              "8        44.538462       34.515385  ...               0.000000        0.000000\n",
              "9        44.250000       34.892857  ...               0.357143        0.178571\n",
              "10       44.704225       33.760563  ...               0.000000        1.000000\n",
              "11       32.000000       36.000000  ...               0.000000        0.000000\n",
              "12       32.400000       19.000000  ...               0.300000        0.000000\n",
              "13       26.000000       13.000000  ...               0.000000        1.000000\n",
              "14       46.505408       36.050147  ...               0.481809        0.000000\n",
              "15       61.000000       54.500000  ...               0.500000        0.500000\n",
              "16       28.000000       14.571429  ...               0.000000        0.000000\n",
              "17       62.000000       56.000000  ...               0.000000        0.000000\n",
              "18       28.833333       18.500000  ...               0.833333        0.000000\n",
              "19       45.746468       35.507064  ...               0.004710        0.000000\n",
              "20       46.204301       36.010753  ...               0.000000        0.000000\n",
              "\n",
              "[21 rows x 20 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 257
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Pqj6TfDeyLWd"
      },
      "source": [
        "У нас получилось 21 отдельных кластеров. \n",
        "Есть интересный кластер(17) где собрались ушедшие клиенты, они все женщины и средний возраст 62 и являются клиентами банка в среднем 56. \n",
        "Но, я решил посмотреть сколько элементов содержится в каждом кластере:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uJDP8Ok8zVsH",
        "outputId": "6b71edb6-faaa-4ac0-b8ca-40bb6de63cb6"
      },
      "source": [
        "# считаем количество элементов\n",
        "res.groupby(\"cluster\")[\"age\"].count()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "cluster\n",
            "0      423\n",
            "1     2519\n",
            "2      471\n",
            "3      376\n",
            "4     1166\n",
            "5     1568\n",
            "6       43\n",
            "7      920\n",
            "8      130\n",
            "9       28\n",
            "10      71\n",
            "11       1\n",
            "12      10\n",
            "13       1\n",
            "14    1017\n",
            "15       2\n",
            "16       7\n",
            "17       1\n",
            "18       6\n",
            "19    1274\n",
            "20      93\n",
            "Name: age, dtype: int64\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3LZiqEih4m6m"
      },
      "source": [
        "Оказалось в 17 кластере 1 элемент. И вообще есть 3 кластера с 1 элементом. И я решил воспользоваться методом DBSCAN который отбрасывает такие данные. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 287
        },
        "id": "FxcGrOZM2gi9",
        "outputId": "edc8f204-769f-4386-f78f-5a689d77cb40"
      },
      "source": [
        "# посчитаем eps\n",
        "from sklearn.neighbors import NearestNeighbors\n",
        "neigh = NearestNeighbors(n_neighbors=2)\n",
        "nbrs = neigh.fit(scaled_bank)\n",
        "distances, indices = nbrs.kneighbors(scaled_bank)\n",
        "distances = np.sort(distances, axis=0)\n",
        "distances = distances[:,1]\n",
        "plt.plot(distances)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[<matplotlib.lines.Line2D at 0x7f7141488d68>]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 260
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAD9CAYAAAC2l2x5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deXRc5Z3m8a8kL9r33ZJsvL02trHxAmZLIBC2iROSDAGarScbgXTSPXM4pxO6G+ikm7gD53SHhBnS4ZAmwCEzzBCTEDqQjcUEiA12wGB+3rXvkrUvpaqaP6okZEW2y1aVqnT1fM7x8a17X1X9Li4ev37ve++bFAwGERER70mOdwEiIhIbCngREY9SwIuIeJQCXkTEoxTwIiIepYAXEfGoOafS2Dl3D3AvsMbM9kw4lg78GNgAjAB3mtlzUapTREROUcQB75xbD2wGqo/T5E6g28yWOueWAa8655aaWW8E7z0f2AQ0Av5IaxIRmeVSgDJgh5kNTTwYUcCHA/gh4AbgpeM0uw64FcDM9jvndgJXAU9H8BGbgFcjqUVERP7MRcD2iTsj7cF/C3jCzI44547Xpopje/c1QGWE798I8OSTT1JaWhrhj4iIzG5NTU3ceOONEM7QiU4a8M6584CNwDeiW9ox/AClpaVUVFTE8GNERDxp0qHtSGbRfBRYCRx2zh0BKoAXnHOXT2hXAywc97oKqD3lMkVEJCpO2oM3s63A1tHX4ZD/xMRZNITG2m8DdoYvsm4iNGYvIiJxMKV58M653c658vDL+4Fc59wB4Dngy2bWM9UCRUTk9JzSPHgAM1s0bnvduO0+4NrolCUiIlOlO1lFRDxKAS8i4lEKeBGROBny+bn9X37LuwfbYvL+CngRkTjp7R+mrqWXhtaTPtHltCjgRUTixO8PrYmdnJQUk/dXwIuIxIk/EAr4lBQFvIiIpwSC4R58cmyiWAEvIhInfn8AgBQN0YiIeEt4hIZkDdGIiHjLaA9eF1lFRDxmdAxeF1lFRDxmdBaNevAiIh4TGJ0mmayAFxHxlLEevAJeRMRjRmfRaIhGRMRbRi+yxijfFfAiIvESHAv42CR8RCs6Oee2AWcAAaAX+JqZ7Z7Q5l7gDqAhvOs1M/tq9EoVEfGWcL7HrAcf6ZJ9t5pZF4Bz7lPAo8D6Sdr9xMzujFZxIiJeNhbwxHEMfjTcw3II9eRFRGQKguGrrEkxGiyPeNFt59wjwOVAEnDlcZpd75y7HGgC7jGz16deooiINwUTZRaNmX3RzKqAu4D7J2nyMHCGmZ0VPv6sc64gOmWKiHjP6CyaWDnlfxiY2ePAJRPD28yazMwX3v41UAusjkqVIiJeFO8evHMu0zlXOe71FqAj/Gt8uwXjttcBiwCLWqUiIh4T+PAqa0xEMgafATztnMsA/ISCfYuZBZ1zzwN3m9lO4D7n3IZwm2HgZjNrik3ZIiIzX6zH4E8a8GbWDGw+zrGrx23fGsW6REQ8L6g7WUVEvOnDG530LBoREU8ZmwevHryIiLcEw7eMxijfFfAiIvHyYQ9eQzQiIp4SGJ1FowU/RES8ZWwWTYzeXwEvIhInY08q0EVWERFvGV10O+4PGxMRkegafVSBxuBFRDxmdAw+RQEvIuIto0M0miYpIuIxGoMXEfEov8bgRUS8KagbnUREvOnDIZrYvL8CXkQkToZ8fgDmzkmJyfsr4EVE4mTY5yclOYm5c2ITxZEs2YdzbhtwBhAAeoGvmdnuCW1SgAeBKwktJbvVzB6JbrkiIt4x5PMzb25seu8QeQ/+VjNba2ZnAw8Aj07S5kZgKbAMOA+41zm3KCpVioh4kM8XYN7c2A2kRPTOZtY17mUOoZ78RNcBPzKzgJm1AtuAa6deooiIN8W6Bx/REA2Ac+4R4HJCzz27cpImVUD1uNc1QOWUqhMR8TDfSIB5MRp/h1O4yGpmXzSzKuAu4P6YVSQiMksMJ8gY/Bgzexy4xDlXMOFQDbBw3OsqoHYKtYmIeNqwz8+8GE2RhAiGaJxzmUCemdWGX28BOsK/xnsa+JJz7hmgALgGuCi65YqIeMfwSCDuY/AZwNPOuQzATyjYt5hZ0Dn3PHC3me0EHgfOBfaHf+5bZnY4FkWLiHjBsM9PVsa8mL3/SQPezJqBzcc5dvW4bT9we/RKExHxttAQTQJcZBURkegaHPaTOi/iyYynTAEvIhInvQM+MtPmxuz9FfAiInHgDwTpH/SRmR67MXgFvIhIHPQP+ggGIStdPXgREU/pG/ABkJ6qgBcR8ZSBoREA0lJ1kVVExFMGh0KLfaRpFo2IiLcMDId78PMV8CIinjIYHqJJnZ9ADxsTEZGpGx2D141OIiIe09M/DBDTZ9Eo4EVE4qC330dyEqRrDF5ExFt6B3xkpM0lOTkpZp+hgBcRiYO+cMDHkgJeRCQOehXwIiLe1DfgIyOGjymAyJbsKyC0WtMSYJjQik23mVnrhHb/AVwGtIV3PW1m/xzVakVEPKJ3YJjKkqyYfkYkl2+DwHfN7CUA59z9wFbgC5O03WpmP4heeSIi3hMMBmntHOBsVxzTz4lkyb4O4KVxu95AS/OJiJy27r5hBof9lOSlx/RzTmkM3jmXTCjcf36cJv/DOfeuc26bc27llKsTEfGg5o5+AEryEyjgge8DvcBkwzB/Byw1szXAM8CvnHOxe8iCiMgMNRrwxYkS8M65B4BlwHVmFph43MzqR/eb2U+ATKAiWoWKiHhFe9cgAIW5aTH9nIgC3jl3H7ABuMbMho7TZsG47SsAP1AfjSJFRLyks3uQuXOSY7rgNkQ2TXIV8E1gH/AH5xzAYTP7tHNuN3C1mTUAjznnSoAA0A180sxGYle6iMjM1NE9SF52KklJsXtMAUQ2i+Y9YNIqzGzduO3LoliXiIhndXQPUpCdGvPP0Z2sIiLTLNSDnx/zz1HAi4hMs87uQfLVgxcR8Za+AR99gyMKeBERr6lu6gagKsbPoQEFvIjItNpXcxSAM8pzYv5ZCngRkWm0c28TVaVZMb+LFRTwIiLTpn/Qx/uHO1gf46dIjlLAi4hMk9+/VYdvJMCFa8un5fMU8CIi0+R3O2tYXJ7D8qq8afk8BbyIyDQY8vk5WNfFhpXFMX9EwSgFvIjINNhtLfgDwWnrvYMCXkRkWjzz0gFK8tNZt7xo2j5TAS8iEmM9/cN8cKSDizdUkDovkqWwo0MBLyISY6/uricQhLOXT8/0yFEKeBGRGPIHgjy3/TBVpVmsXJQ/rZ+tgBcRiaHf7aihtrmH6y5bTnLy9MyeGaWAFxGJkUAgyLOvHGRRWTYXrVtw8h+IskiW7CsAHgeWAMPAfuA2M2ud0C4d+DGhtVtHgDvN7LmoVywiMkPseL+J6qYe/vq6ddM29328SHrwQeC7ZubMbA1wENg6Sbs7gW4zWwpsAR5xzmVGr1QRkZmjf9DHj597j8LcNC7ZUBmXGk4a8GbWYWYvjdv1BrBwkqbXAT8M/8x+YCdwVRRqFBGZcX71+hHqW/v4m+vPJiUlPqPhp/Spzrlk4Hbg55McrgKqx72uAeLz15aISJy9/HY9KxbmsXbZ9N3YNNGp/rXyfaAX+EEMahER8YTa5h4ONXSxeXVZXOuIOOCdcw8Ay4DrzCwwSZMajh26qQJqp1aeiMjM88Ib1aQkJ3Hppqq41hFRwDvn7iM0O+YaMxs6TrOngdvC7ZcBm4BfRaNIEZGZwu8P8LudtWxeU0Zu1vy41nLSgHfOrQK+CZQDf3DO7XbO/Sx8bLdzbvTJ9fcDuc65A8BzwJfNrCdGdYuIJKT3j3TQ0z/MRWunf977RCedB29m7wGTTuA0s3XjtvuAa6NXmojIzPPmnibmpCRztovfxdVRupNVRCRK+gd9vPjmEda7YtJT58a7HAW8iEi0vG0tDAz5uer8RfEuBVDAi4hERTAY5NmXD1KYmzati3qciAJeRCQKtu9u4IPqTq7/+HLmxOnO1YkSowoRkRlu+zv1ZKXP47JzJnuSS3wo4EVEpmj3vhb+8E4jl59bRco0P/P9RBTwIiJTEAwG+fdt75KXNZ/PXbY83uUcQwEvIjIF7x5so7a5l+s+7hJiauR4CngRkdMUDAZ5/Pm9ZKXP42MbE+/huQp4EZHTtGNvMx9Ud3LzVStIm3/SBwNMOwW8iMhp8AeCPLJtD+WFGQk1c2Y8BbyIyGl4Y08jje193PJfzmTunMSM0sSsSkQkgY2/azXei3qciAJeROQU/b/fH2DvkQ4+ffGShJr3PpECXkTkFDR39PPkr/Zywdpytly4ON7lnJACXkTkFDz9230EAkG++MnVJCUlbu8dFPAiIhF7dXc9L7xRzdUXnEFhblq8yzmpiCZuhhfc/iywCFhjZnsmaXMvcAfQEN71mpl9NTpliojE1/uH2/nXp95m5aJ8Pr9lVbzLiUikM/O3Ad8DXj1Ju5+Y2Z1TK0lEJLEMDo/wvZ/uIit9Lnf95TnMnZMS75IiEtEQjZltN7PaWBcjIpJogsEgD/7v3TS29/HX168nN2t+vEuKWLTvrb3eOXc50ATcY2avR/n9RUSm1VMvGq/urueWq1ey3hXHu5xTEs2LrA8DZ5jZWcD9wLPOuYIovr+IyLQZ9vn5923v8tSLxmWbqvjsJcviXdIpi1rAm1mTmfnC278GaoHV0Xp/EZHpEgwG+f7Tu/nFq4e46rxF/NXn1pGcwDc0HU/UAt45t2Dc9jpCM24sWu8vIjJdfrH9EC+9Vce1ly7jjv+6NqHvVj2RSKdJPgh8BigFfuOcazezVc6554G7zWwncJ9zbgPgB4aBm82sKVaFi4jEwgfVHTzy7B6WVeZy05Ur413OlEQU8Gb2deDrk+y/etz2rVGsS0Rk2jW19/Gd//gjxXnp/OOXz5uRwzLj6U5WERGgpbOfu/7XawwN+7nrL88hK31evEuassRbgkREZJrtOdjGvzy+k6FhP/fdcQGLF+TEu6SoUMCLyKzlG/Hzo217eOGNI5QVZvBPt53PwrLseJcVNQp4EZmV/P4A3/vpbl7eVceWixbzF5c7Mj0wLDOeAl5EZh2/P8C3H32Ttz5o4S+uWMENl7t4lxQTCngRmVX6Bnxs/ckOdu9r5UufWs2WixJ70Y6pUMCLyKxR29zDA0++RXVjN1/4ZCjcE33RjqlQwIuI5wUCQZ7bfojHfvk+8+fN4a7/dg7nnFka77JiTgEvIp7mGwnwncf+yI73m9l0Zglfu3Ydedmp8S5rWijgRcSzevqH+fuH/8Ch+i6+dM1qtlzo7SGZiRTwIuJJ+2s7+d5Pd1HX0ss3bt3EBWeVx7ukaaeAFxFPae8a4In//IDf7awhNyuVv//8uWxcWRLvsuJCAS8inhAMBvnZSwd48gUjEAjwiQsXc/3lzhPPlDldCngRmdH8gSB/+FMDz7y0nwN1XWxeXcoXPrma0oKMeJcWdwp4EZmxOnsG+bendvG2tVBWkMHfXH82H9tYOasupJ6IAl5EZpzOnkEef34vv91RA0lJ3PbpNVx9/hkz/vnt0aaAF5EZo2/Ax7aXD7Lt5QP4RgJccd4itly4mMqSrHiXlpBOGvDOuQeAzxJaY3WNme2ZpE0K8CBwJRAEtprZI9EtVURmq87uQX7+6iF++dohBob8XHBWOTddtYKKYgX7iUTSg98GfA949QRtbgSWAsuAAmCXc+43ZnZkyhWKyKw04g/w+juNvLK7jp17W/AHAly4dgGfuWQpSyty413ejHDSgDez7QDOnfBxmtcBPzKzANDqnNsGXAvcH40iRWT22FfTyYtvVvPKrnoGhkbIz07l6vMXccXmhVSVemcxjukQrTH4KqB63OsaoDJK7y0iHjfiD/DKrnpefLOa9w61M3dOMh85ewHnn1XOxhUlunh6mnSRVUTiIhAIcqi+izf2NPLim9V09gxRVpjB57es4orNC0lPnRvvEme8aAV8DbAQ2BF+PbFHLyLCsM/P3sMd/P7tWnbubaardxiAs5cX8VefW8ymlSWawx5F0Qr4p4EvOeeeIXSR9Rrgoii9t4jMYP5AkD0H29j+pwZefruOgaER0lPnsGllKetXFHG2KyYva3Y8vne6RTJN8kHgM0Ap8BvnXLuZrXLOPQ/cbWY7gceBc4H94R/7lpkdjlXRIpLYfCN+3j/cwY73m3n57TqO9g6ROi+FzavLuGBtOeuWFZE6XyPEsRbJLJqvA1+fZP/V47b9wO3RLU1EZgp/IMjBuqPsOdjGnkPtvHOgjaFhP3NSktiwooRLNlayYUUxqfMU6tNJ/7VF5LR09w3zzoFWXtlVzzv7W+kbHAGgrDCDyzZVsd4Vs2ZpIWnqqceN/suLSMQ6uwfZsbeZ199t5O0PmgkEISdzHhesXcDaZYWsWVqo8fQEooAXkePyB4IcbujirQ+aeWNPEwdqjwJQlJfGpy9eyubVZSytzGVOSnKcK5XJKOBFZIw/EKS2uYe9RzrYZS28e6CN3gEfAEsrc7npqhVsXFHC4gU5ms44AyjgRWYx30iAfTWd7NrXgh3pxGo6GRgKjaXnZ8/nvDVlnLU0NPRSkJMW52rlVCngRWaR0UD/4EgH7xxow2o66RvwkZwEC8uyuXhDBSsW5rNiUR5lBRnqpc9wCngRjxrxBzjS2E1Tex/1rb3srznKn/a3MjjsB6CyJJMLzirnbFfEuuXFZKbp0QBeo4AX8YBgMEhTez/vHmyjrqWX9w+3c7ihm2Gff6xNcV4al2ysZN2yIs48o4DcrPlxrFimgwJeZIbp7R+muqmH+tbesd75vupO2roGAZiTksyyylyuPG8hK6ryqSjJpKwgQ3eOzkL6ExdJYMFg6ImL7x/uGJvdcqSxe+x4SnISxfnprFiUz+olhZy1tJDyokxS9HhdQQEvkjB8IwEaWsPDK43d1Lf0UtfSQ0f3EAAZqXNYVpXHTetWsLg8h8qSLApz0zQHXY5LAS8yzQKBIK1HB6ht7qGpvQ+r6WR/zVEa2/sIBIIAZKTNpaIokzVLili3vIizXRH52ama1SKnRAEvEiO9/cPUtfZS39JLY1sfDW19NLT1UtfSy9Dwhxc/s9LnsXpJAeefVUZVSRbLq/IoK9QURZk6BbxIFHT3DXOg9ihW3cHu/a3UNvfQ0+8bO56cBCX5GZQVZbDqjAIqijOpKs2mrDCDvKz5CnOJCQW8yCkIBoN09gxR29TD4cYu9tccZV9tJ03t/WNtllTkcMHaBZTmp1NZksWC4kyK89KZO0dj5TK9FPAiEwSDQbr7hmntHKC5o5/mjj4a2/s5XN9FTXPP2K38AIW5aSyrzOWKzYtYVpnL0opcMnTDkCSIiALeObcceIzQcnztwC1mtn9Cm3uBO4CG8K7XzOyr0StVJPpGL3hWN3azPzzEcqiha2yt0FGZaXM5ozyHSzdWUl6USVVpFlUlWeRl69G4krgi7cE/DDxkZk84524Cfgh8bJJ2PzGzO6NWnUgU+EYCtHcN0N41SH1rL7XNoZuEGlr7aO3sZ3gkAEBSEiwszeacM0upKs2mJD997Jd65TITRbImazGwHvh4eNdTwA+cc0Vm1hrL4kQiMezz09LZT0vHAM2d/bR29tPY1kd71yBdvUM0dfSPTT8EmDsnmfLCDKpKs9h0ZkmoR16SxZIFObrbUzwlkm9zJVAfXncVM/M75xrC+ycG/PXOucuBJuAeM3s9qtXKrNbTP8yhui7q23qpbeqhsb2PhtY+mjr6CH6Y36QkJ1GUl0ZxXjpnLMjhwnWhC54FuWmUFqRTkp+hOz1lVohmd+Vh4J/NzOec+zjwrHNupZm1R/EzxMN8IwFaj/bT3N5PS2c/rZ0DdHQP0tjeR11LL0d7hsbaps1Poawwk8ULcrh4QwWlBRkU56VRkp9Bfk6qAlyEyAK+FljgnEsJ995TgPLw/jFm1jRu+9fOuVpgNfByNAuWma+3f5iGtj5qmnpoaAvdBHSksZuG1l7GjaSQnAQ5mfMpLchg08oSFhSFAr2qNIu8rFSSFeIiJ3TSgDezFufcbuAG4Inw77smjr875xaYWX14ex2wCLCoVywzgj8QpKNrkLqW0FBKTVPPWIh3juuJJycnUZyXxsLSbM4/q5yygnRKCjIoyUsnPydVz1kRmYJIh2i+AjzmnLsb6ARuAXDOPQ/cbWY7gfuccxsAPzAM3Dy+Vy/e4xsJ0NrZT1N7aK54XWuoN97Y1kdTex8j/g+742nzU1hUlsOGFSVUFGdSUZxJeVGmnnwoEkMRBbyZfQCcO8n+q8dt3xrFuiQBBINBjvYMjQV4U0c/Te19NHeEQr29a+CYi5vz5qZQXphBZUkW564qpaQgg4qiTMoKMyjI0YOyRKab5oTNcoNDI9S29NDSOUDb0YFwePeFQ73/mBWBAPKzUynJT2fNkgJKCzLG5omXFmSQn61xcZFEooCfJYZ8fmqbe6hr6aU6vE7n6IXOEX9grF3a/DmUFqSzoCiDDSuKx8K7JD+d4vx05s9NieNZiMipUMB7UHvXAFbdyb6azrGFI1o6+8eGU0ZXASoryGDdRUWsWJRHSX5oGCU7Y56GUkQ8QgE/A/lGAjS09VLT1ENjWx+d3YN09AzS2T1EY3vf2HzxOSlJVJZk4RbmcenGSqpKs6koDo2Jz1NPXMTzFPAJaNjnp61rgPqWXlqPhsbGW48O0H50kPau0Di5f9yE8YzUOeRlp5KXlcqGFcUsLs9h+cI8lizIYe4cBbnIbKWAn2bDPj81zT20Hx3gaO8QHV2DtHcP0t41OPZArO6+Y59kmJKcREFOKgU5aSypyOWCteVUlWRRVZpNeVEGqfP0xygif07JECMDQyNjt9s3tYfmhteHF1QeGDp2Zkpu5nzyc1IpzE3DLcynICeVwpxUFhRlUVKQTk7mfM0VF5FTpoCfIt9IgD++38R7h9pp7eynpXOA1s7+Y5ZrA0idl0JpQQYfXV/J2mWFFOelk5eVSm7WfK30IyIxoYA/BYPDI7R3DdLc3s/+uk7eP9yBVXfSN+AjbX4KRXnpFOel46ryxp5mWJyXTmlBOrlad1NEppkCfhIDQyM0tvVxuKGL/bVHqW3uoaGtj7ajA8e0qyjO5Pw1ZWxeU8aGFSUaRhGRhDJrA3503c228CyV5s5+9h7uYF9NJy2dHwZ56rwUFpVls3pJARVFmRTlpVOUm8ai8myy0ufF8QxERE5s1gV8bXMPb+xp5Lnth+noHjzmWF7WfNYsKeSKzaHZKQtLs/UwLBGZsTwd8MFgkJqmHv50oJX3DrVzsK6L5o5+AFYtLuAzlyylKDeNgpxUivPTyc3UOLmIeIfnAv71dxt5871Gmtr7qWvpoas3NKe8OD+dZRW5fPIji9m8uozivPQ4VyoiElueCvhfbj/Ewz97l8y0uSwsy2bjyhJWLy7krKWFFOcr0EVkdvFMwP/xvSYe/cV7rFyUz3fuuIAUrQQkIrOcJ1Jw7+EOvv3om+TnpPK3t2xUuIuIEGEP3jm3HHgMKADagVvMbP+ENinAg8CVQBDYamaPRLfcyT35wl6y0ufxb//9YjLS5k7HR4qIJLxIu7oPAw+Z2XLgIeCHk7S5EVgKLAPOA+51zi2KRpEn0tjWx5/2t3HppkqFu4jIOCcNeOdcMbAeeCq86ylgvXOuaELT64AfmVnAzFqBbcC10Sx2MgfqjgJw0boFsf4oEZEZJZIefCVQb2Z+gPDvDeH941UB1eNe10zSJuqGfX7mzUlmQVFmrD9KRGRGmfGzaC7dVMU5q0o1PCMiMkEkPfhaYEH4IuroxdTy8P7xaoCF415XTdImJvRMGBGRP3fSgDezFmA3cEN41w3ArvA4+3hPA19yziWHx+evAf5vNIsVEZHIRTqL5ivA15xz+4CvhV/jnHveObcx3OZx4BCwH3gD+JaZHY5yvSIiEqGIxuDN7APg3En2Xz1u2w/cHr3SRERkKnTLp4iIRyngRUQ8SgEvIuJRiTIPPgWgqakp3nWIiMwY4zIzZbLjiRLwZQA33nhjvOsQEZmJyoCDE3cmSsDvAC4CGgF/nGsREZkpUgiF+47JDiYFg8HpLUdERKaFLrKKiHiUAl5ExKMU8CIiHqWAFxHxKAW8iIhHKeBFRDxKAS8i4lGJcqPTaXPOLQceAwqAduAWM9sf36oi55wrIPQs/SXAMKHn6d9mZq3Ouc3AD4E04AhwU3gBFk73WCJxzt0D3AusMbM9Xj1f51wq8K/AZcAg8LqZfflE393TPZYonHOfAL4NJIV//aOZPeOVc3bOPQB8FlhE+Pt7sjrjce5e6ME/DDxkZsuBhwj9jz6TBIHvmpkzszWEbjfe6pxLBp4Avho+t1eArQCneyyROOfWA5sJL9Tu8fP9LqFgXx7+M/6H8P4TfXdP91jcOeeSCHVabjazdcDNwGPhPyuvnPM24COEv7/jxOL8TvvcZ3TAO+eKgfXAU+FdTwHrw0sGzghm1mFmL43b9QahtW03AINmtj28/2Hgc+Ht0z2WEJxz8wl9UccvEOPJ83XOZQK3AP9gZkEAM2s+0Xf3dI9NzxlFLADkhLdzCT2GpBCPnLOZbTezY9acjsWf6VTPfUYHPFAJ1IdXkxpdVaohvH/GCfdwbgd+TmjR8rHegZm1AcnOufwpHEsU3wKeMLMj4/Z59XyXEPpn9T3OuZ3OuZeccxdy4u/u6R5LCOG/yD4HPOucqybU270FD59zWCzOb0rnPtMD3mu+D/QCP4h3IbHinDsP2Aj8z3jXMk1SgMWEFqrfCPwt8AyQGdeqYsg5Nwf4JvApM1sIbAH+Dx4+50Q10wO+FljgnEsBCP9eHt4/o4Qv2iwDrjOzAFBDaKhm9HghEDCzjikcSwQfBVYCh51zR4AK4AVgKd483xpghPA/sc3sTaANGOD4390Tfa9nwnd+HVBuZq8BhH/vI3QdwqvnDKd/DjE79xkd8OGZEruBG8K7biDUU2qNX1Wnzjl3H6Gx5GvMbCi8+y0gLfzPeYCvAE9P8VjcmdlWMys3s0VmtgioA64A7seb59sG/B74OIzNiCgG9nGc7+6Jvtcz5DtfB1Q45xyAc24lUEJohphXz/mEeRSLY5HUNOMfF+ycW0FoClEe0EloCpHFt6rIOedWAQ813roAAACcSURBVHsI/Q8/EN592Mw+7Zw7n9AV81Q+nP7XHP650zqWaMK9+E+Ep0l68nydc4uBRwlNc/MBf2dm/3mi7+7pHksUzrkbgW8QutgKcI+ZbfPKOTvnHgQ+A5QS+hdZu5mtisX5TeXcZ3zAi4jI5Gb0EI2IiByfAl5ExKMU8CIiHqWAFxHxKAW8iIhHKeBFRDxKAS8i4lEKeBERj/r/unAthx0dIEgAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 582
        },
        "id": "py9-Omao2idj",
        "outputId": "a636fddd-2cce-41ca-dd56-00a4062f562d"
      },
      "source": [
        "dbmodel = DBSCAN(eps=2.9, min_samples=6)\n",
        "dbmodel = dbmodel.fit(scaled_bank)\n",
        "dbpred = dbmodel.labels_\n",
        "\n",
        "res = bank.copy()\n",
        "res[\"cluster\"] = dbmodel.labels_\n",
        "res.groupby(\"cluster\").mean()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>age</th>\n",
              "      <th>months_on_book</th>\n",
              "      <th>credit_limit</th>\n",
              "      <th>total_trans_ct</th>\n",
              "      <th>att_flag_Existing Customer</th>\n",
              "      <th>gender_M</th>\n",
              "      <th>edu_level_Doctorate</th>\n",
              "      <th>edu_level_Graduate</th>\n",
              "      <th>edu_level_High School</th>\n",
              "      <th>edu_level_Post-Graduate</th>\n",
              "      <th>edu_level_Uneducated</th>\n",
              "      <th>edu_level_Unknown</th>\n",
              "      <th>marital_status_Married</th>\n",
              "      <th>marital_status_Single</th>\n",
              "      <th>marital_status_Unknown</th>\n",
              "      <th>income_$40K - $60K</th>\n",
              "      <th>income_$60K - $80K</th>\n",
              "      <th>income_$80K - $120K</th>\n",
              "      <th>income_Less than $40K</th>\n",
              "      <th>income_Unknown</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>cluster</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>-1</th>\n",
              "      <td>44.655556</td>\n",
              "      <td>34.122222</td>\n",
              "      <td>13612.986667</td>\n",
              "      <td>59.555556</td>\n",
              "      <td>0.500000</td>\n",
              "      <td>0.644444</td>\n",
              "      <td>0.266667</td>\n",
              "      <td>0.022222</td>\n",
              "      <td>0.033333</td>\n",
              "      <td>0.366667</td>\n",
              "      <td>0.155556</td>\n",
              "      <td>0.133333</td>\n",
              "      <td>0.100000</td>\n",
              "      <td>0.055556</td>\n",
              "      <td>0.822222</td>\n",
              "      <td>0.188889</td>\n",
              "      <td>0.166667</td>\n",
              "      <td>0.144444</td>\n",
              "      <td>0.088889</td>\n",
              "      <td>0.266667</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>46.382906</td>\n",
              "      <td>35.992723</td>\n",
              "      <td>8476.431569</td>\n",
              "      <td>64.539428</td>\n",
              "      <td>0.844536</td>\n",
              "      <td>0.522228</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.342816</td>\n",
              "      <td>0.218709</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.159301</td>\n",
              "      <td>0.167637</td>\n",
              "      <td>0.503043</td>\n",
              "      <td>0.419291</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.196613</td>\n",
              "      <td>0.155200</td>\n",
              "      <td>0.169092</td>\n",
              "      <td>0.397327</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>45.734280</td>\n",
              "      <td>35.367140</td>\n",
              "      <td>8731.366329</td>\n",
              "      <td>68.434077</td>\n",
              "      <td>0.866126</td>\n",
              "      <td>0.452333</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.344828</td>\n",
              "      <td>0.227181</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.152130</td>\n",
              "      <td>0.174442</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.235294</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.233266</td>\n",
              "      <td>0.456389</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>46.134831</td>\n",
              "      <td>36.101124</td>\n",
              "      <td>10962.111236</td>\n",
              "      <td>62.011236</td>\n",
              "      <td>0.887640</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.370787</td>\n",
              "      <td>0.247191</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.146067</td>\n",
              "      <td>0.123596</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>45.435028</td>\n",
              "      <td>35.186441</td>\n",
              "      <td>7884.603672</td>\n",
              "      <td>65.293785</td>\n",
              "      <td>0.836158</td>\n",
              "      <td>0.420904</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.536723</td>\n",
              "      <td>0.392655</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.293785</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.186441</td>\n",
              "      <td>0.449153</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>46.371336</td>\n",
              "      <td>35.832790</td>\n",
              "      <td>9352.082519</td>\n",
              "      <td>66.079262</td>\n",
              "      <td>0.834962</td>\n",
              "      <td>0.045603</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.336591</td>\n",
              "      <td>0.222584</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.188925</td>\n",
              "      <td>0.148751</td>\n",
              "      <td>0.473398</td>\n",
              "      <td>0.437568</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>46.862745</td>\n",
              "      <td>36.568627</td>\n",
              "      <td>10341.450980</td>\n",
              "      <td>60.294118</td>\n",
              "      <td>0.921569</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.431373</td>\n",
              "      <td>0.509804</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>47.349835</td>\n",
              "      <td>36.702970</td>\n",
              "      <td>7590.307921</td>\n",
              "      <td>64.821782</td>\n",
              "      <td>0.778878</td>\n",
              "      <td>0.402640</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.481848</td>\n",
              "      <td>0.429043</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.221122</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.181518</td>\n",
              "      <td>0.488449</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>45.486486</td>\n",
              "      <td>36.243243</td>\n",
              "      <td>10433.251351</td>\n",
              "      <td>71.567568</td>\n",
              "      <td>0.972973</td>\n",
              "      <td>0.027027</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.486486</td>\n",
              "      <td>0.405405</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>48.265625</td>\n",
              "      <td>38.171875</td>\n",
              "      <td>9553.848438</td>\n",
              "      <td>61.687500</td>\n",
              "      <td>0.781250</td>\n",
              "      <td>0.031250</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.531250</td>\n",
              "      <td>0.375000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>46.445946</td>\n",
              "      <td>36.162162</td>\n",
              "      <td>10658.913514</td>\n",
              "      <td>62.445946</td>\n",
              "      <td>0.783784</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.405405</td>\n",
              "      <td>0.432432</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>47.375000</td>\n",
              "      <td>34.625000</td>\n",
              "      <td>16305.500000</td>\n",
              "      <td>46.625000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>45.000000</td>\n",
              "      <td>34.000000</td>\n",
              "      <td>9144.613636</td>\n",
              "      <td>72.515152</td>\n",
              "      <td>0.848485</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.333333</td>\n",
              "      <td>0.272727</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.106061</td>\n",
              "      <td>0.090909</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>44.400000</td>\n",
              "      <td>35.200000</td>\n",
              "      <td>3961.030000</td>\n",
              "      <td>68.000000</td>\n",
              "      <td>0.900000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>42.000000</td>\n",
              "      <td>35.222222</td>\n",
              "      <td>6003.888889</td>\n",
              "      <td>74.777778</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.333333</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.888889</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "               age  months_on_book  ...  income_Less than $40K  income_Unknown\n",
              "cluster                             ...                                       \n",
              "-1       44.655556       34.122222  ...               0.088889        0.266667\n",
              " 0       46.382906       35.992723  ...               0.397327        0.000000\n",
              " 1       45.734280       35.367140  ...               0.456389        0.000000\n",
              " 2       46.134831       36.101124  ...               0.000000        0.000000\n",
              " 3       45.435028       35.186441  ...               0.449153        0.000000\n",
              " 4       46.371336       35.832790  ...               0.000000        1.000000\n",
              " 5       46.862745       36.568627  ...               0.000000        0.000000\n",
              " 6       47.349835       36.702970  ...               0.488449        0.000000\n",
              " 7       45.486486       36.243243  ...               0.000000        1.000000\n",
              " 8       48.265625       38.171875  ...               0.000000        1.000000\n",
              " 9       46.445946       36.162162  ...               0.000000        0.000000\n",
              " 10      47.375000       34.625000  ...               0.000000        0.000000\n",
              " 11      45.000000       34.000000  ...               0.000000        1.000000\n",
              " 12      44.400000       35.200000  ...               1.000000        0.000000\n",
              " 13      42.000000       35.222222  ...               0.888889        0.000000\n",
              "\n",
              "[15 rows x 20 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 263
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "JKzbRIkk21XS",
        "outputId": "c5888592-1b7b-4c54-9137-0da4619663cf"
      },
      "source": [
        "res.groupby(\"cluster\")[\"age\"].count()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "cluster\n",
              "-1       90\n",
              " 0     7558\n",
              " 1      493\n",
              " 2       89\n",
              " 3      354\n",
              " 4      921\n",
              " 5       51\n",
              " 6      303\n",
              " 7       37\n",
              " 8       64\n",
              " 9       74\n",
              " 10       8\n",
              " 11      66\n",
              " 12      10\n",
              " 13       9\n",
              "Name: age, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 264
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mI1eG4zM6SdW"
      },
      "source": [
        "Получилось 14 кластеров. 1 большой и 13 маленьких. \n",
        "\n",
        "Можем видеть 3 кластера(13,11,7) с наибольшем количеством транзакции и там больше всего женщины.\n",
        "\n",
        "Еще есть интересная небольшая группа(10) которая состоит из ушедших клиентов, и они все мужчины, средний возраст которых 47 лет, они все магистранты или аспиранты по образованию и у них самый высокий кредитный лимит. Можно подумать над тем, что для этих людей размер кредитного лимита был не важным показателем при выборе банка. \n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IBfXf8uM9gUf"
      },
      "source": [
        "## Ассоциативные правила"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6IiaZCa4B4hJ"
      },
      "source": [
        "Найдем закономерности на основе характеристик клиентов"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "idm2AkVhBCWh",
        "outputId": "efba0c1f-92a2-49dc-dca4-dc7e4f3f1c7a"
      },
      "source": [
        "# оставляем нужные нам данные\n",
        "bank_cat = pd.read_csv(\"BankChurners.csv\")\n",
        "bank_cat = bank_cat[['Attrition_Flag', 'Customer_Age', 'Gender', 'Marital_Status', 'Income_Category', 'Months_on_book', 'Credit_Limit', 'Total_Trans_Ct']]\n",
        "bank_cat.columns = ['att_flag', 'age', 'gender', 'marital_status', 'income', 'months_on_book', 'credit_limit', 'total_trans_ct']\n",
        "bank_cat.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>att_flag</th>\n",
              "      <th>age</th>\n",
              "      <th>gender</th>\n",
              "      <th>marital_status</th>\n",
              "      <th>income</th>\n",
              "      <th>months_on_book</th>\n",
              "      <th>credit_limit</th>\n",
              "      <th>total_trans_ct</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>45</td>\n",
              "      <td>M</td>\n",
              "      <td>Married</td>\n",
              "      <td>$60K - $80K</td>\n",
              "      <td>39</td>\n",
              "      <td>12691.0</td>\n",
              "      <td>42</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>49</td>\n",
              "      <td>F</td>\n",
              "      <td>Single</td>\n",
              "      <td>Less than $40K</td>\n",
              "      <td>44</td>\n",
              "      <td>8256.0</td>\n",
              "      <td>33</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>51</td>\n",
              "      <td>M</td>\n",
              "      <td>Married</td>\n",
              "      <td>$80K - $120K</td>\n",
              "      <td>36</td>\n",
              "      <td>3418.0</td>\n",
              "      <td>20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>40</td>\n",
              "      <td>F</td>\n",
              "      <td>Unknown</td>\n",
              "      <td>Less than $40K</td>\n",
              "      <td>34</td>\n",
              "      <td>3313.0</td>\n",
              "      <td>20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>40</td>\n",
              "      <td>M</td>\n",
              "      <td>Married</td>\n",
              "      <td>$60K - $80K</td>\n",
              "      <td>21</td>\n",
              "      <td>4716.0</td>\n",
              "      <td>28</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "            att_flag  age gender  ... months_on_book credit_limit  total_trans_ct\n",
              "0  Existing Customer   45      M  ...             39      12691.0              42\n",
              "1  Existing Customer   49      F  ...             44       8256.0              33\n",
              "2  Existing Customer   51      M  ...             36       3418.0              20\n",
              "3  Existing Customer   40      F  ...             34       3313.0              20\n",
              "4  Existing Customer   40      M  ...             21       4716.0              28\n",
              "\n",
              "[5 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 282
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "go-_d4-1CVVs",
        "outputId": "737398d9-5dcf-4fcb-b4b3-41605493bb40"
      },
      "source": [
        "# числовые переменные переобразуем на категоральные\n",
        "bank_cat[\"age\"] = pd.cut(bank_cat[\"age\"], bins = [18, 30, 50, 70, 100], right = False)\n",
        "bank_cat[\"months_on_book\"] = pd.cut(bank_cat[\"months_on_book\"], bins= 3, right = False)\n",
        "bank_cat[\"credit_limit\"] = pd.cut(bank_cat[\"credit_limit\"], bins= 6, right = False)\n",
        "bank_cat[\"total_trans_ct\"] = pd.cut(bank_cat[\"total_trans_ct\"], bins= 6, right = False)\n",
        "bank_cat.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>att_flag</th>\n",
              "      <th>age</th>\n",
              "      <th>gender</th>\n",
              "      <th>marital_status</th>\n",
              "      <th>income</th>\n",
              "      <th>months_on_book</th>\n",
              "      <th>credit_limit</th>\n",
              "      <th>total_trans_ct</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>[30, 50)</td>\n",
              "      <td>M</td>\n",
              "      <td>Married</td>\n",
              "      <td>$60K - $80K</td>\n",
              "      <td>[27.333, 41.667)</td>\n",
              "      <td>[12464.2, 17977.15)</td>\n",
              "      <td>[31.5, 53.0)</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>[30, 50)</td>\n",
              "      <td>F</td>\n",
              "      <td>Single</td>\n",
              "      <td>Less than $40K</td>\n",
              "      <td>[41.667, 56.043)</td>\n",
              "      <td>[6951.25, 12464.2)</td>\n",
              "      <td>[31.5, 53.0)</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>[50, 70)</td>\n",
              "      <td>M</td>\n",
              "      <td>Married</td>\n",
              "      <td>$80K - $120K</td>\n",
              "      <td>[27.333, 41.667)</td>\n",
              "      <td>[1438.3, 6951.25)</td>\n",
              "      <td>[10.0, 31.5)</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>[30, 50)</td>\n",
              "      <td>F</td>\n",
              "      <td>Unknown</td>\n",
              "      <td>Less than $40K</td>\n",
              "      <td>[27.333, 41.667)</td>\n",
              "      <td>[1438.3, 6951.25)</td>\n",
              "      <td>[10.0, 31.5)</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Existing Customer</td>\n",
              "      <td>[30, 50)</td>\n",
              "      <td>M</td>\n",
              "      <td>Married</td>\n",
              "      <td>$60K - $80K</td>\n",
              "      <td>[13.0, 27.333)</td>\n",
              "      <td>[1438.3, 6951.25)</td>\n",
              "      <td>[10.0, 31.5)</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "            att_flag       age  ...         credit_limit total_trans_ct\n",
              "0  Existing Customer  [30, 50)  ...  [12464.2, 17977.15)   [31.5, 53.0)\n",
              "1  Existing Customer  [30, 50)  ...   [6951.25, 12464.2)   [31.5, 53.0)\n",
              "2  Existing Customer  [50, 70)  ...    [1438.3, 6951.25)   [10.0, 31.5)\n",
              "3  Existing Customer  [30, 50)  ...    [1438.3, 6951.25)   [10.0, 31.5)\n",
              "4  Existing Customer  [30, 50)  ...    [1438.3, 6951.25)   [10.0, 31.5)\n",
              "\n",
              "[5 rows x 8 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 283
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yVG81y1uDVDE",
        "outputId": "cbb876a5-e2c1-4475-a69b-c9a6ef44516f"
      },
      "source": [
        "!pip install mlxtend"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Requirement already satisfied: mlxtend in /usr/local/lib/python3.6/dist-packages (0.14.0)\n",
            "Requirement already satisfied: scikit-learn>=0.18 in /usr/local/lib/python3.6/dist-packages (from mlxtend) (0.22.2.post1)\n",
            "Requirement already satisfied: scipy>=0.17 in /usr/local/lib/python3.6/dist-packages (from mlxtend) (1.4.1)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.6/dist-packages (from mlxtend) (50.3.2)\n",
            "Requirement already satisfied: matplotlib>=1.5.1 in /usr/local/lib/python3.6/dist-packages (from mlxtend) (3.2.2)\n",
            "Requirement already satisfied: pandas>=0.17.1 in /usr/local/lib/python3.6/dist-packages (from mlxtend) (1.1.4)\n",
            "Requirement already satisfied: numpy>=1.10.4 in /usr/local/lib/python3.6/dist-packages (from mlxtend) (1.18.5)\n",
            "Requirement already satisfied: joblib>=0.11 in /usr/local/lib/python3.6/dist-packages (from scikit-learn>=0.18->mlxtend) (0.17.0)\n",
            "Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=1.5.1->mlxtend) (2.4.7)\n",
            "Requirement already satisfied: python-dateutil>=2.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=1.5.1->mlxtend) (2.8.1)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=1.5.1->mlxtend) (1.3.1)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=1.5.1->mlxtend) (0.10.0)\n",
            "Requirement already satisfied: pytz>=2017.2 in /usr/local/lib/python3.6/dist-packages (from pandas>=0.17.1->mlxtend) (2018.9)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.6/dist-packages (from python-dateutil>=2.1->matplotlib>=1.5.1->mlxtend) (1.15.0)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 241
        },
        "id": "RyC1GHunDXJ5",
        "outputId": "be38aa07-68c6-43e3-84d3-05235e7957b7"
      },
      "source": [
        "# категоральные переобразуем на числовые\n",
        "bank_cat = pd.get_dummies(bank_cat)\n",
        "bank_cat.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>att_flag_Attrited Customer</th>\n",
              "      <th>att_flag_Existing Customer</th>\n",
              "      <th>age_[18, 30)</th>\n",
              "      <th>age_[30, 50)</th>\n",
              "      <th>age_[50, 70)</th>\n",
              "      <th>age_[70, 100)</th>\n",
              "      <th>gender_F</th>\n",
              "      <th>gender_M</th>\n",
              "      <th>marital_status_Divorced</th>\n",
              "      <th>marital_status_Married</th>\n",
              "      <th>marital_status_Single</th>\n",
              "      <th>marital_status_Unknown</th>\n",
              "      <th>income_$120K +</th>\n",
              "      <th>income_$40K - $60K</th>\n",
              "      <th>income_$60K - $80K</th>\n",
              "      <th>income_$80K - $120K</th>\n",
              "      <th>income_Less than $40K</th>\n",
              "      <th>income_Unknown</th>\n",
              "      <th>months_on_book_[13.0, 27.333)</th>\n",
              "      <th>months_on_book_[27.333, 41.667)</th>\n",
              "      <th>months_on_book_[41.667, 56.043)</th>\n",
              "      <th>credit_limit_[1438.3, 6951.25)</th>\n",
              "      <th>credit_limit_[6951.25, 12464.2)</th>\n",
              "      <th>credit_limit_[12464.2, 17977.15)</th>\n",
              "      <th>credit_limit_[17977.15, 23490.1)</th>\n",
              "      <th>credit_limit_[23490.1, 29003.05)</th>\n",
              "      <th>credit_limit_[29003.05, 34549.078)</th>\n",
              "      <th>total_trans_ct_[10.0, 31.5)</th>\n",
              "      <th>total_trans_ct_[31.5, 53.0)</th>\n",
              "      <th>total_trans_ct_[53.0, 74.5)</th>\n",
              "      <th>total_trans_ct_[74.5, 96.0)</th>\n",
              "      <th>total_trans_ct_[96.0, 117.5)</th>\n",
              "      <th>total_trans_ct_[117.5, 139.129)</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   att_flag_Attrited Customer  ...  total_trans_ct_[117.5, 139.129)\n",
              "0                           0  ...                                0\n",
              "1                           0  ...                                0\n",
              "2                           0  ...                                0\n",
              "3                           0  ...                                0\n",
              "4                           0  ...                                0\n",
              "\n",
              "[5 rows x 33 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 284
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 204
        },
        "id": "W_TziYrpDvgz",
        "outputId": "2622d404-3e4a-4b36-dca4-be93dee1dccf"
      },
      "source": [
        "from mlxtend.frequent_patterns import apriori\n",
        "frequent_itemsets = apriori(bank_cat, min_support=0.1, use_colnames=True)\n",
        "frequent_itemsets['length'] = frequent_itemsets['itemsets'].apply(lambda x: len(x))\n",
        "frequent_itemsets.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>support</th>\n",
              "      <th>itemsets</th>\n",
              "      <th>length</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.160660</td>\n",
              "      <td>(att_flag_Attrited Customer)</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.839340</td>\n",
              "      <td>(att_flag_Existing Customer)</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.632171</td>\n",
              "      <td>(age_[30, 50))</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.348376</td>\n",
              "      <td>(age_[50, 70))</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.529081</td>\n",
              "      <td>(gender_F)</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    support                      itemsets  length\n",
              "0  0.160660  (att_flag_Attrited Customer)       1\n",
              "1  0.839340  (att_flag_Existing Customer)       1\n",
              "2  0.632171                (age_[30, 50))       1\n",
              "3  0.348376                (age_[50, 70))       1\n",
              "4  0.529081                    (gender_F)       1"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 289
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "FzrSdLp6Eiz1",
        "outputId": "3389f509-3ffe-4138-d5f8-74a53e67797c"
      },
      "source": [
        "# фильтруем полученные закономерности\n",
        "rules = association_rules(frequent_itemsets, metric=\"lift\", min_threshold=2.1)\n",
        "rules[ (rules['lift'] >= 2.1) &\n",
        "       (rules['confidence'] >= 0.4) ]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>antecedents</th>\n",
              "      <th>consequents</th>\n",
              "      <th>antecedent support</th>\n",
              "      <th>consequent support</th>\n",
              "      <th>support</th>\n",
              "      <th>confidence</th>\n",
              "      <th>lift</th>\n",
              "      <th>leverage</th>\n",
              "      <th>conviction</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>(total_trans_ct_[31.5, 53.0))</td>\n",
              "      <td>(att_flag_Attrited Customer)</td>\n",
              "      <td>0.245482</td>\n",
              "      <td>0.160660</td>\n",
              "      <td>0.100820</td>\n",
              "      <td>0.410700</td>\n",
              "      <td>2.556336</td>\n",
              "      <td>0.061380</td>\n",
              "      <td>1.424300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>(att_flag_Attrited Customer)</td>\n",
              "      <td>(total_trans_ct_[31.5, 53.0))</td>\n",
              "      <td>0.160660</td>\n",
              "      <td>0.245482</td>\n",
              "      <td>0.100820</td>\n",
              "      <td>0.627535</td>\n",
              "      <td>2.556336</td>\n",
              "      <td>0.061380</td>\n",
              "      <td>2.025743</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>(age_[50, 70))</td>\n",
              "      <td>(months_on_book_[41.667, 56.043))</td>\n",
              "      <td>0.348376</td>\n",
              "      <td>0.215859</td>\n",
              "      <td>0.191370</td>\n",
              "      <td>0.549320</td>\n",
              "      <td>2.544813</td>\n",
              "      <td>0.116170</td>\n",
              "      <td>1.739906</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>(months_on_book_[41.667, 56.043))</td>\n",
              "      <td>(age_[50, 70))</td>\n",
              "      <td>0.215859</td>\n",
              "      <td>0.348376</td>\n",
              "      <td>0.191370</td>\n",
              "      <td>0.886551</td>\n",
              "      <td>2.544813</td>\n",
              "      <td>0.116170</td>\n",
              "      <td>5.743754</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>(income_$60K - $80K)</td>\n",
              "      <td>(gender_M)</td>\n",
              "      <td>0.138442</td>\n",
              "      <td>0.470919</td>\n",
              "      <td>0.138442</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>2.123506</td>\n",
              "      <td>0.073247</td>\n",
              "      <td>inf</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>(income_$80K - $120K)</td>\n",
              "      <td>(gender_M)</td>\n",
              "      <td>0.151575</td>\n",
              "      <td>0.470919</td>\n",
              "      <td>0.151575</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>2.123506</td>\n",
              "      <td>0.080195</td>\n",
              "      <td>inf</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>(months_on_book_[41.667, 56.043), att_flag_Exi...</td>\n",
              "      <td>(age_[50, 70))</td>\n",
              "      <td>0.181199</td>\n",
              "      <td>0.348376</td>\n",
              "      <td>0.160660</td>\n",
              "      <td>0.886649</td>\n",
              "      <td>2.545093</td>\n",
              "      <td>0.097534</td>\n",
              "      <td>5.748705</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>(age_[50, 70), att_flag_Existing Customer)</td>\n",
              "      <td>(months_on_book_[41.667, 56.043))</td>\n",
              "      <td>0.291399</td>\n",
              "      <td>0.215859</td>\n",
              "      <td>0.160660</td>\n",
              "      <td>0.551339</td>\n",
              "      <td>2.554165</td>\n",
              "      <td>0.097759</td>\n",
              "      <td>1.747735</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>(months_on_book_[41.667, 56.043))</td>\n",
              "      <td>(age_[50, 70), att_flag_Existing Customer)</td>\n",
              "      <td>0.215859</td>\n",
              "      <td>0.291399</td>\n",
              "      <td>0.160660</td>\n",
              "      <td>0.744282</td>\n",
              "      <td>2.554165</td>\n",
              "      <td>0.097759</td>\n",
              "      <td>2.771022</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>(age_[50, 70))</td>\n",
              "      <td>(months_on_book_[41.667, 56.043), att_flag_Exi...</td>\n",
              "      <td>0.348376</td>\n",
              "      <td>0.181199</td>\n",
              "      <td>0.160660</td>\n",
              "      <td>0.461168</td>\n",
              "      <td>2.545093</td>\n",
              "      <td>0.097534</td>\n",
              "      <td>1.519585</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>(income_$60K - $80K, att_flag_Existing Customer)</td>\n",
              "      <td>(gender_M)</td>\n",
              "      <td>0.119779</td>\n",
              "      <td>0.470919</td>\n",
              "      <td>0.119779</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>2.123506</td>\n",
              "      <td>0.063373</td>\n",
              "      <td>inf</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>(income_$60K - $80K)</td>\n",
              "      <td>(gender_M, att_flag_Existing Customer)</td>\n",
              "      <td>0.138442</td>\n",
              "      <td>0.402093</td>\n",
              "      <td>0.119779</td>\n",
              "      <td>0.865193</td>\n",
              "      <td>2.151720</td>\n",
              "      <td>0.064112</td>\n",
              "      <td>4.435265</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>(att_flag_Existing Customer, income_$80K - $120K)</td>\n",
              "      <td>(gender_M)</td>\n",
              "      <td>0.127678</td>\n",
              "      <td>0.470919</td>\n",
              "      <td>0.127678</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>2.123506</td>\n",
              "      <td>0.067552</td>\n",
              "      <td>inf</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>(age_[50, 70), gender_F)</td>\n",
              "      <td>(months_on_book_[41.667, 56.043))</td>\n",
              "      <td>0.187519</td>\n",
              "      <td>0.215859</td>\n",
              "      <td>0.100326</td>\n",
              "      <td>0.535018</td>\n",
              "      <td>2.478560</td>\n",
              "      <td>0.059848</td>\n",
              "      <td>1.686392</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>(months_on_book_[41.667, 56.043), gender_F)</td>\n",
              "      <td>(age_[50, 70))</td>\n",
              "      <td>0.113360</td>\n",
              "      <td>0.348376</td>\n",
              "      <td>0.100326</td>\n",
              "      <td>0.885017</td>\n",
              "      <td>2.540411</td>\n",
              "      <td>0.060834</td>\n",
              "      <td>5.667157</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21</th>\n",
              "      <td>(months_on_book_[41.667, 56.043))</td>\n",
              "      <td>(age_[50, 70), gender_F)</td>\n",
              "      <td>0.215859</td>\n",
              "      <td>0.187519</td>\n",
              "      <td>0.100326</td>\n",
              "      <td>0.464776</td>\n",
              "      <td>2.478560</td>\n",
              "      <td>0.059848</td>\n",
              "      <td>1.518021</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>(age_[50, 70), credit_limit_[1438.3, 6951.25))</td>\n",
              "      <td>(months_on_book_[41.667, 56.043))</td>\n",
              "      <td>0.220993</td>\n",
              "      <td>0.215859</td>\n",
              "      <td>0.122840</td>\n",
              "      <td>0.555853</td>\n",
              "      <td>2.575081</td>\n",
              "      <td>0.075137</td>\n",
              "      <td>1.765501</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23</th>\n",
              "      <td>(months_on_book_[41.667, 56.043), credit_limit...</td>\n",
              "      <td>(age_[50, 70))</td>\n",
              "      <td>0.137652</td>\n",
              "      <td>0.348376</td>\n",
              "      <td>0.122840</td>\n",
              "      <td>0.892396</td>\n",
              "      <td>2.561591</td>\n",
              "      <td>0.074885</td>\n",
              "      <td>6.055762</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>(months_on_book_[41.667, 56.043))</td>\n",
              "      <td>(age_[50, 70), credit_limit_[1438.3, 6951.25))</td>\n",
              "      <td>0.215859</td>\n",
              "      <td>0.220993</td>\n",
              "      <td>0.122840</td>\n",
              "      <td>0.569076</td>\n",
              "      <td>2.575081</td>\n",
              "      <td>0.075137</td>\n",
              "      <td>1.807758</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>27</th>\n",
              "      <td>(months_on_book_[27.333, 41.667), income_$80K ...</td>\n",
              "      <td>(gender_M)</td>\n",
              "      <td>0.102696</td>\n",
              "      <td>0.470919</td>\n",
              "      <td>0.102696</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>2.123506</td>\n",
              "      <td>0.054334</td>\n",
              "      <td>inf</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29</th>\n",
              "      <td>(income_$80K - $120K)</td>\n",
              "      <td>(months_on_book_[27.333, 41.667), gender_M)</td>\n",
              "      <td>0.151575</td>\n",
              "      <td>0.301570</td>\n",
              "      <td>0.102696</td>\n",
              "      <td>0.677524</td>\n",
              "      <td>2.246657</td>\n",
              "      <td>0.056985</td>\n",
              "      <td>2.165838</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>30</th>\n",
              "      <td>(months_on_book_[41.667, 56.043), att_flag_Exi...</td>\n",
              "      <td>(age_[50, 70))</td>\n",
              "      <td>0.115730</td>\n",
              "      <td>0.348376</td>\n",
              "      <td>0.103782</td>\n",
              "      <td>0.896758</td>\n",
              "      <td>2.574111</td>\n",
              "      <td>0.063464</td>\n",
              "      <td>6.311601</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>31</th>\n",
              "      <td>(age_[50, 70), att_flag_Existing Customer, cre...</td>\n",
              "      <td>(months_on_book_[41.667, 56.043))</td>\n",
              "      <td>0.185247</td>\n",
              "      <td>0.215859</td>\n",
              "      <td>0.103782</td>\n",
              "      <td>0.560235</td>\n",
              "      <td>2.595377</td>\n",
              "      <td>0.063795</td>\n",
              "      <td>1.783090</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32</th>\n",
              "      <td>(months_on_book_[41.667, 56.043), att_flag_Exi...</td>\n",
              "      <td>(age_[50, 70), credit_limit_[1438.3, 6951.25))</td>\n",
              "      <td>0.181199</td>\n",
              "      <td>0.220993</td>\n",
              "      <td>0.103782</td>\n",
              "      <td>0.572752</td>\n",
              "      <td>2.591716</td>\n",
              "      <td>0.063738</td>\n",
              "      <td>1.823313</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>33</th>\n",
              "      <td>(months_on_book_[41.667, 56.043), credit_limit...</td>\n",
              "      <td>(age_[50, 70), att_flag_Existing Customer)</td>\n",
              "      <td>0.137652</td>\n",
              "      <td>0.291399</td>\n",
              "      <td>0.103782</td>\n",
              "      <td>0.753945</td>\n",
              "      <td>2.587328</td>\n",
              "      <td>0.063670</td>\n",
              "      <td>2.879853</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35</th>\n",
              "      <td>(age_[50, 70), credit_limit_[1438.3, 6951.25))</td>\n",
              "      <td>(months_on_book_[41.667, 56.043), att_flag_Exi...</td>\n",
              "      <td>0.220993</td>\n",
              "      <td>0.181199</td>\n",
              "      <td>0.103782</td>\n",
              "      <td>0.469616</td>\n",
              "      <td>2.591716</td>\n",
              "      <td>0.063738</td>\n",
              "      <td>1.543789</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>36</th>\n",
              "      <td>(months_on_book_[41.667, 56.043))</td>\n",
              "      <td>(age_[50, 70), att_flag_Existing Customer, cre...</td>\n",
              "      <td>0.215859</td>\n",
              "      <td>0.185247</td>\n",
              "      <td>0.103782</td>\n",
              "      <td>0.480787</td>\n",
              "      <td>2.595377</td>\n",
              "      <td>0.063795</td>\n",
              "      <td>1.569206</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>38</th>\n",
              "      <td>(total_trans_ct_[74.5, 96.0), income_Less than...</td>\n",
              "      <td>(att_flag_Existing Customer, gender_F)</td>\n",
              "      <td>0.113360</td>\n",
              "      <td>0.437247</td>\n",
              "      <td>0.106843</td>\n",
              "      <td>0.942509</td>\n",
              "      <td>2.155552</td>\n",
              "      <td>0.057277</td>\n",
              "      <td>9.788492</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>40</th>\n",
              "      <td>(months_on_book_[27.333, 41.667), gender_F, cr...</td>\n",
              "      <td>(age_[30, 50), income_Less than $40K)</td>\n",
              "      <td>0.273131</td>\n",
              "      <td>0.218821</td>\n",
              "      <td>0.136269</td>\n",
              "      <td>0.498915</td>\n",
              "      <td>2.280016</td>\n",
              "      <td>0.076503</td>\n",
              "      <td>1.558976</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>41</th>\n",
              "      <td>(age_[30, 50), gender_F, credit_limit_[1438.3,...</td>\n",
              "      <td>(months_on_book_[27.333, 41.667), income_Less ...</td>\n",
              "      <td>0.264145</td>\n",
              "      <td>0.225141</td>\n",
              "      <td>0.136269</td>\n",
              "      <td>0.515888</td>\n",
              "      <td>2.291402</td>\n",
              "      <td>0.076800</td>\n",
              "      <td>1.600578</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>42</th>\n",
              "      <td>(months_on_book_[27.333, 41.667), income_Less ...</td>\n",
              "      <td>(age_[30, 50), gender_F, credit_limit_[1438.3,...</td>\n",
              "      <td>0.225141</td>\n",
              "      <td>0.264145</td>\n",
              "      <td>0.136269</td>\n",
              "      <td>0.605263</td>\n",
              "      <td>2.291402</td>\n",
              "      <td>0.076800</td>\n",
              "      <td>1.864165</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>43</th>\n",
              "      <td>(age_[30, 50), income_Less than $40K)</td>\n",
              "      <td>(months_on_book_[27.333, 41.667), gender_F, cr...</td>\n",
              "      <td>0.218821</td>\n",
              "      <td>0.273131</td>\n",
              "      <td>0.136269</td>\n",
              "      <td>0.622744</td>\n",
              "      <td>2.280016</td>\n",
              "      <td>0.076503</td>\n",
              "      <td>1.926724</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>44</th>\n",
              "      <td>(months_on_book_[27.333, 41.667), att_flag_Exi...</td>\n",
              "      <td>(age_[30, 50), income_Less than $40K)</td>\n",
              "      <td>0.222080</td>\n",
              "      <td>0.218821</td>\n",
              "      <td>0.110694</td>\n",
              "      <td>0.498444</td>\n",
              "      <td>2.277861</td>\n",
              "      <td>0.062099</td>\n",
              "      <td>1.557510</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>45</th>\n",
              "      <td>(age_[30, 50), att_flag_Existing Customer, gen...</td>\n",
              "      <td>(months_on_book_[27.333, 41.667), income_Less ...</td>\n",
              "      <td>0.215266</td>\n",
              "      <td>0.225141</td>\n",
              "      <td>0.110694</td>\n",
              "      <td>0.514220</td>\n",
              "      <td>2.283995</td>\n",
              "      <td>0.062229</td>\n",
              "      <td>1.595083</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>46</th>\n",
              "      <td>(months_on_book_[27.333, 41.667), att_flag_Exi...</td>\n",
              "      <td>(age_[30, 50), gender_F, credit_limit_[1438.3,...</td>\n",
              "      <td>0.183964</td>\n",
              "      <td>0.264145</td>\n",
              "      <td>0.110694</td>\n",
              "      <td>0.601718</td>\n",
              "      <td>2.277979</td>\n",
              "      <td>0.062101</td>\n",
              "      <td>1.847570</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>47</th>\n",
              "      <td>(months_on_book_[27.333, 41.667), gender_F, cr...</td>\n",
              "      <td>(age_[30, 50), att_flag_Existing Customer, inc...</td>\n",
              "      <td>0.273131</td>\n",
              "      <td>0.180113</td>\n",
              "      <td>0.110694</td>\n",
              "      <td>0.405278</td>\n",
              "      <td>2.250139</td>\n",
              "      <td>0.061500</td>\n",
              "      <td>1.378607</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>48</th>\n",
              "      <td>(age_[30, 50), att_flag_Existing Customer, inc...</td>\n",
              "      <td>(months_on_book_[27.333, 41.667), gender_F, cr...</td>\n",
              "      <td>0.180113</td>\n",
              "      <td>0.273131</td>\n",
              "      <td>0.110694</td>\n",
              "      <td>0.614583</td>\n",
              "      <td>2.250139</td>\n",
              "      <td>0.061500</td>\n",
              "      <td>1.885930</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49</th>\n",
              "      <td>(age_[30, 50), gender_F, credit_limit_[1438.3,...</td>\n",
              "      <td>(months_on_book_[27.333, 41.667), att_flag_Exi...</td>\n",
              "      <td>0.264145</td>\n",
              "      <td>0.183964</td>\n",
              "      <td>0.110694</td>\n",
              "      <td>0.419065</td>\n",
              "      <td>2.277979</td>\n",
              "      <td>0.062101</td>\n",
              "      <td>1.404696</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50</th>\n",
              "      <td>(months_on_book_[27.333, 41.667), income_Less ...</td>\n",
              "      <td>(age_[30, 50), att_flag_Existing Customer, gen...</td>\n",
              "      <td>0.225141</td>\n",
              "      <td>0.215266</td>\n",
              "      <td>0.110694</td>\n",
              "      <td>0.491667</td>\n",
              "      <td>2.283995</td>\n",
              "      <td>0.062229</td>\n",
              "      <td>1.543739</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>51</th>\n",
              "      <td>(age_[30, 50), income_Less than $40K)</td>\n",
              "      <td>(months_on_book_[27.333, 41.667), att_flag_Exi...</td>\n",
              "      <td>0.218821</td>\n",
              "      <td>0.222080</td>\n",
              "      <td>0.110694</td>\n",
              "      <td>0.505866</td>\n",
              "      <td>2.277861</td>\n",
              "      <td>0.062099</td>\n",
              "      <td>1.574312</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                          antecedents  ... conviction\n",
              "0                       (total_trans_ct_[31.5, 53.0))  ...   1.424300\n",
              "1                        (att_flag_Attrited Customer)  ...   2.025743\n",
              "2                                      (age_[50, 70))  ...   1.739906\n",
              "3                   (months_on_book_[41.667, 56.043))  ...   5.743754\n",
              "4                                (income_$60K - $80K)  ...        inf\n",
              "7                               (income_$80K - $120K)  ...        inf\n",
              "8   (months_on_book_[41.667, 56.043), att_flag_Exi...  ...   5.748705\n",
              "9          (age_[50, 70), att_flag_Existing Customer)  ...   1.747735\n",
              "10                  (months_on_book_[41.667, 56.043))  ...   2.771022\n",
              "11                                     (age_[50, 70))  ...   1.519585\n",
              "12   (income_$60K - $80K, att_flag_Existing Customer)  ...        inf\n",
              "14                               (income_$60K - $80K)  ...   4.435265\n",
              "16  (att_flag_Existing Customer, income_$80K - $120K)  ...        inf\n",
              "18                           (age_[50, 70), gender_F)  ...   1.686392\n",
              "19        (months_on_book_[41.667, 56.043), gender_F)  ...   5.667157\n",
              "21                  (months_on_book_[41.667, 56.043))  ...   1.518021\n",
              "22     (age_[50, 70), credit_limit_[1438.3, 6951.25))  ...   1.765501\n",
              "23  (months_on_book_[41.667, 56.043), credit_limit...  ...   6.055762\n",
              "25                  (months_on_book_[41.667, 56.043))  ...   1.807758\n",
              "27  (months_on_book_[27.333, 41.667), income_$80K ...  ...        inf\n",
              "29                              (income_$80K - $120K)  ...   2.165838\n",
              "30  (months_on_book_[41.667, 56.043), att_flag_Exi...  ...   6.311601\n",
              "31  (age_[50, 70), att_flag_Existing Customer, cre...  ...   1.783090\n",
              "32  (months_on_book_[41.667, 56.043), att_flag_Exi...  ...   1.823313\n",
              "33  (months_on_book_[41.667, 56.043), credit_limit...  ...   2.879853\n",
              "35     (age_[50, 70), credit_limit_[1438.3, 6951.25))  ...   1.543789\n",
              "36                  (months_on_book_[41.667, 56.043))  ...   1.569206\n",
              "38  (total_trans_ct_[74.5, 96.0), income_Less than...  ...   9.788492\n",
              "40  (months_on_book_[27.333, 41.667), gender_F, cr...  ...   1.558976\n",
              "41  (age_[30, 50), gender_F, credit_limit_[1438.3,...  ...   1.600578\n",
              "42  (months_on_book_[27.333, 41.667), income_Less ...  ...   1.864165\n",
              "43              (age_[30, 50), income_Less than $40K)  ...   1.926724\n",
              "44  (months_on_book_[27.333, 41.667), att_flag_Exi...  ...   1.557510\n",
              "45  (age_[30, 50), att_flag_Existing Customer, gen...  ...   1.595083\n",
              "46  (months_on_book_[27.333, 41.667), att_flag_Exi...  ...   1.847570\n",
              "47  (months_on_book_[27.333, 41.667), gender_F, cr...  ...   1.378607\n",
              "48  (age_[30, 50), att_flag_Existing Customer, inc...  ...   1.885930\n",
              "49  (age_[30, 50), gender_F, credit_limit_[1438.3,...  ...   1.404696\n",
              "50  (months_on_book_[27.333, 41.667), income_Less ...  ...   1.543739\n",
              "51              (age_[30, 50), income_Less than $40K)  ...   1.574312\n",
              "\n",
              "[40 rows x 9 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 290
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NgrP5V_KAAWU"
      },
      "source": [
        "Можно отметить закономерность, если у клиента за последние 12месяцев количество транзакций было между 31-53 то клиент уходил. \n",
        "\n",
        "В остальном получилось достаточно банальные закономерности, связанное с возрастом и периодом отношение с банком. Такие данные не очень помогут при решении главной задачи. "
      ]
    }
  ]
}