fraud-modeling-CDR-K-Means.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.3"
    },
    "colab": {
      "name": "K-Means.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/multimediary/portfolio-data-science/blob/main/fraud-modeling-CDR-K-Means.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "u4iY8H7VwvFa"
      },
      "source": [
        "**1. IMPORT LIBRARY**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_h-zjM9KaxLi",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a7532d94-5cd4-4442-a945-9d17ea2921d9"
      },
      "source": [
        "# For ignore warning\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')\n",
        "warnings.simplefilter('ignore')\n",
        "\n",
        "# Library For code execution time\n",
        "!pip install ipython-autotime\n",
        "%load_ext autotime\n",
        "\n",
        "# Library For Preprocessing\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "from sklearn.preprocessing import MinMaxScaler, LabelEncoder, StandardScaler\n",
        "\n",
        "# Library For Clustering\n",
        "from sklearn.cluster import KMeans\n",
        "\n",
        "# Library For Visualization\n",
        "import plotly\n",
        "import plotly.graph_objs as go\n",
        "from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot\n",
        "from sklearn.decomposition import PCA\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Library For Evaluation\n",
        "from sklearn.metrics import confusion_matrix, accuracy_score, precision_score, recall_score, f1_score"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting ipython-autotime\n",
            "  Downloading ipython_autotime-0.3.1-py2.py3-none-any.whl (6.8 kB)\n",
            "Requirement already satisfied: ipython in /usr/local/lib/python3.7/dist-packages (from ipython-autotime) (5.5.0)\n",
            "Requirement already satisfied: decorator in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (4.4.2)\n",
            "Requirement already satisfied: traitlets>=4.2 in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (5.1.0)\n",
            "Requirement already satisfied: prompt-toolkit<2.0.0,>=1.0.4 in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (1.0.18)\n",
            "Requirement already satisfied: setuptools>=18.5 in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (57.4.0)\n",
            "Requirement already satisfied: pickleshare in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (0.7.5)\n",
            "Requirement already satisfied: simplegeneric>0.8 in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (0.8.1)\n",
            "Requirement already satisfied: pexpect in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (4.8.0)\n",
            "Requirement already satisfied: pygments in /usr/local/lib/python3.7/dist-packages (from ipython->ipython-autotime) (2.6.1)\n",
            "Requirement already satisfied: six>=1.9.0 in /usr/local/lib/python3.7/dist-packages (from prompt-toolkit<2.0.0,>=1.0.4->ipython->ipython-autotime) (1.15.0)\n",
            "Requirement already satisfied: wcwidth in /usr/local/lib/python3.7/dist-packages (from prompt-toolkit<2.0.0,>=1.0.4->ipython->ipython-autotime) (0.2.5)\n",
            "Requirement already satisfied: ptyprocess>=0.5 in /usr/local/lib/python3.7/dist-packages (from pexpect->ipython->ipython-autotime) (0.7.0)\n",
            "Installing collected packages: ipython-autotime\n",
            "Successfully installed ipython-autotime-0.3.1\n",
            "time: 1.65 s (started: 2021-10-11 14:32:17 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "x_XuR48GpqP9"
      },
      "source": [
        "**2. DATA GATHERING**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FY8kYBHywWfO",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 220
        },
        "outputId": "c073d5a9-8d4b-44ce-e4f6-e49c5b9f07dd"
      },
      "source": [
        "# Read Dataset with Pandas from Git\n",
        "df = pd.read_csv(\"https://github.com/multimediary/dataset/blob/master/dataset.csv?raw=true\")\n",
        "df.head()"
      ],
      "execution_count": 2,
      "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>starttime</th>\n",
              "      <th>stoptime</th>\n",
              "      <th>anumber</th>\n",
              "      <th>bnumber</th>\n",
              "      <th>duration</th>\n",
              "      <th>fee</th>\n",
              "      <th>destination</th>\n",
              "      <th>type</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>2018-05-02 09:13:24</td>\n",
              "      <td>2018-05-02 09:20:38</td>\n",
              "      <td>62626292960</td>\n",
              "      <td>1500200</td>\n",
              "      <td>434</td>\n",
              "      <td>6000</td>\n",
              "      <td>callcenter</td>\n",
              "      <td>Injapati</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>2018-05-18 16:49:19</td>\n",
              "      <td>2018-05-18 16:52:11</td>\n",
              "      <td>62626292960</td>\n",
              "      <td>1500200</td>\n",
              "      <td>172</td>\n",
              "      <td>2250</td>\n",
              "      <td>callcenter</td>\n",
              "      <td>Injapati</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2018-05-18 18:15:17</td>\n",
              "      <td>2018-05-18 18:18:20</td>\n",
              "      <td>62626292960</td>\n",
              "      <td>1500200</td>\n",
              "      <td>183</td>\n",
              "      <td>3000</td>\n",
              "      <td>callcenter</td>\n",
              "      <td>Injapati</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>2018-05-31 08:52:46</td>\n",
              "      <td>2018-05-31 08:53:03</td>\n",
              "      <td>62626292253</td>\n",
              "      <td>1500788</td>\n",
              "      <td>17</td>\n",
              "      <td>750</td>\n",
              "      <td>callcenter</td>\n",
              "      <td>Injapati</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>2018-05-31 08:53:10</td>\n",
              "      <td>2018-05-31 08:53:24</td>\n",
              "      <td>62626292253</td>\n",
              "      <td>1500788</td>\n",
              "      <td>14</td>\n",
              "      <td>750</td>\n",
              "      <td>callcenter</td>\n",
              "      <td>Injapati</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             starttime             stoptime  ...      type label\n",
              "0  2018-05-02 09:13:24  2018-05-02 09:20:38  ...  Injapati     0\n",
              "1  2018-05-18 16:49:19  2018-05-18 16:52:11  ...  Injapati     0\n",
              "2  2018-05-18 18:15:17  2018-05-18 18:18:20  ...  Injapati     0\n",
              "3  2018-05-31 08:52:46  2018-05-31 08:53:03  ...  Injapati     0\n",
              "4  2018-05-31 08:53:10  2018-05-31 08:53:24  ...  Injapati     0\n",
              "\n",
              "[5 rows x 9 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 2
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 1.04 s (started: 2021-10-11 14:32:19 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vj6eSmf5yOSR",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "381c8bf3-f75a-4a5d-a542-deef7055ad22"
      },
      "source": [
        "# Show Dataset Description (Focus at Dtype)\n",
        "df.info()"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 11418 entries, 0 to 11417\n",
            "Data columns (total 9 columns):\n",
            " #   Column       Non-Null Count  Dtype \n",
            "---  ------       --------------  ----- \n",
            " 0   starttime    11418 non-null  object\n",
            " 1   stoptime     11418 non-null  object\n",
            " 2   anumber      11418 non-null  int64 \n",
            " 3   bnumber      11418 non-null  object\n",
            " 4   duration     11418 non-null  int64 \n",
            " 5   fee          11418 non-null  int64 \n",
            " 6   destination  11418 non-null  object\n",
            " 7   type         11418 non-null  object\n",
            " 8   label        11418 non-null  int64 \n",
            "dtypes: int64(4), object(5)\n",
            "memory usage: 803.0+ KB\n",
            "time: 16.1 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "atMYvtJYC9t5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        },
        "outputId": "5ee0ccff-7c28-4bc4-a211-0ec95c3136b8"
      },
      "source": [
        "# Displays Dataframe Statistics (Focus at std)\n",
        "df.describe()"
      ],
      "execution_count": 4,
      "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>anumber</th>\n",
              "      <th>duration</th>\n",
              "      <th>fee</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>1.141800e+04</td>\n",
              "      <td>11418.000000</td>\n",
              "      <td>1.141800e+04</td>\n",
              "      <td>11418.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>6.262625e+10</td>\n",
              "      <td>154.111841</td>\n",
              "      <td>1.215145e+04</td>\n",
              "      <td>0.034770</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>3.985148e+04</td>\n",
              "      <td>410.204092</td>\n",
              "      <td>4.131463e+04</td>\n",
              "      <td>0.183204</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>6.262621e+10</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>4.300000e+01</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>6.262621e+10</td>\n",
              "      <td>24.000000</td>\n",
              "      <td>6.340000e+02</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>6.262621e+10</td>\n",
              "      <td>57.000000</td>\n",
              "      <td>2.614000e+03</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>6.262629e+10</td>\n",
              "      <td>143.000000</td>\n",
              "      <td>1.045400e+04</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>6.262630e+10</td>\n",
              "      <td>17735.000000</td>\n",
              "      <td>1.567841e+06</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "            anumber      duration           fee         label\n",
              "count  1.141800e+04  11418.000000  1.141800e+04  11418.000000\n",
              "mean   6.262625e+10    154.111841  1.215145e+04      0.034770\n",
              "std    3.985148e+04    410.204092  4.131463e+04      0.183204\n",
              "min    6.262621e+10      0.000000  4.300000e+01      0.000000\n",
              "25%    6.262621e+10     24.000000  6.340000e+02      0.000000\n",
              "50%    6.262621e+10     57.000000  2.614000e+03      0.000000\n",
              "75%    6.262629e+10    143.000000  1.045400e+04      0.000000\n",
              "max    6.262630e+10  17735.000000  1.567841e+06      1.000000"
            ]
          },
          "metadata": {},
          "execution_count": 4
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 26.7 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ERzzPo6swj60",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "408d3428-2b86-4f08-c188-5a790bebc5f2"
      },
      "source": [
        "# Show dataset shape\n",
        "df.shape"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(11418, 9)"
            ]
          },
          "metadata": {},
          "execution_count": 5
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 3.41 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5NcmExGDwpe7",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "34a33d04-cf25-4747-807c-e1a4ec3ff0e3"
      },
      "source": [
        "# Show the dataset labeled fraud (1) and non fraud (0)\n",
        "df['label'].value_counts()"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    11021\n",
              "1      397\n",
              "Name: label, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 6
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 5.6 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gdASqke5xCat",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "38c03f8b-c786-4515-814d-ac236ec56270"
      },
      "source": [
        "# Show a comparison between the dataset labeled fraud and non-fraud\n",
        "print(df['label'].value_counts() / df['label'].shape[0])"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0    0.96523\n",
            "1    0.03477\n",
            "Name: label, dtype: float64\n",
            "time: 19.6 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "idKX9pCW01jm",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "00fc520c-debe-41ea-9c74-b902d1f98dd4"
      },
      "source": [
        "# Check what types of calls are there\n",
        "df['type'].value_counts()"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "SLJJ          7093\n",
              "SLI 001       4035\n",
              "Lokal          153\n",
              "SLI 007        102\n",
              "VoIP 01017      21\n",
              "Injapati        14\n",
              "Name: type, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 8
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 7.25 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CLcq74ZVT67o",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "9e887cc6-d1e4-4b2e-a1d9-fa42e81064da"
      },
      "source": [
        "# Find null in the dataset\n",
        "df.isnull().sum().max()"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0"
            ]
          },
          "metadata": {},
          "execution_count": 9
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 10.4 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fr55APKM1y_2"
      },
      "source": [
        "**3. DATA PREPARATION**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BM8j8Rkb2cI9",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "08665a1e-ac50-4c1e-88f9-31c21963b278"
      },
      "source": [
        "# Drop unneeded attributes, namely startime and stoptime, destination and label\n",
        "dataset = df.drop(['starttime','stoptime','destination','label'], axis = 1)\n",
        "label = df.label"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 3.61 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wNs92sk74z5l",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "b5cfeaff-536d-4f4e-d565-9970941c1f4b"
      },
      "source": [
        "enc = LabelEncoder()\n",
        "def transforming(input, j):\n",
        "    input.iloc[:,j] = enc.fit_transform(input.iloc[:,j])\n",
        "\n",
        "# Change data in the form of objects / strings into numbers\n",
        "for i in range(0,5):\n",
        "    if(i!=2 or i!=3):\n",
        "        transforming(dataset, i)"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 27 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ucktkXe951aG",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 220
        },
        "outputId": "9dd1c73f-7b51-4d4e-a638-1de0ee090738"
      },
      "source": [
        "# Show Dataset \n",
        "dataset.head()"
      ],
      "execution_count": 12,
      "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>anumber</th>\n",
              "      <th>bnumber</th>\n",
              "      <th>duration</th>\n",
              "      <th>fee</th>\n",
              "      <th>type</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>265</td>\n",
              "      <td>1333</td>\n",
              "      <td>431</td>\n",
              "      <td>447</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>265</td>\n",
              "      <td>1333</td>\n",
              "      <td>172</td>\n",
              "      <td>238</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>265</td>\n",
              "      <td>1333</td>\n",
              "      <td>183</td>\n",
              "      <td>294</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>225</td>\n",
              "      <td>1334</td>\n",
              "      <td>17</td>\n",
              "      <td>96</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>225</td>\n",
              "      <td>1334</td>\n",
              "      <td>14</td>\n",
              "      <td>96</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   anumber  bnumber  duration  fee  type\n",
              "0      265     1333       431  447     0\n",
              "1      265     1333       172  238     0\n",
              "2      265     1333       183  294     0\n",
              "3      225     1334        17   96     0\n",
              "4      225     1334        14   96     0"
            ]
          },
          "metadata": {},
          "execution_count": 12
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 14.3 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VFwFuKZB59bd",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "1b0cc7f8-08bf-4c42-fc66-75ac4ea3a687"
      },
      "source": [
        "# Transform data with Scaler to perform data normalization so that the data used does not have large deviations.\n",
        "features = ['anumber', 'bnumber', 'duration', 'fee', 'type']\n",
        "Final_Dataset = dataset.loc[:, features].values\n",
        "Final_Dataset = StandardScaler().fit_transform(Final_Dataset)\n",
        "Final_Dataset = pd.DataFrame(data = Final_Dataset, columns = features)"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 8.71 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yHUS7rgl6c53",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 220
        },
        "outputId": "ccb50a4c-639e-432f-802c-6fd592757bca"
      },
      "source": [
        "# Show the Final Dataset \n",
        "Final_Dataset.head()"
      ],
      "execution_count": 14,
      "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>anumber</th>\n",
              "      <th>bnumber</th>\n",
              "      <th>duration</th>\n",
              "      <th>fee</th>\n",
              "      <th>type</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1.097724</td>\n",
              "      <td>-0.319205</td>\n",
              "      <td>1.697394</td>\n",
              "      <td>0.358047</td>\n",
              "      <td>-3.259819</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1.097724</td>\n",
              "      <td>-0.319205</td>\n",
              "      <td>0.250598</td>\n",
              "      <td>-0.376284</td>\n",
              "      <td>-3.259819</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1.097724</td>\n",
              "      <td>-0.319205</td>\n",
              "      <td>0.312045</td>\n",
              "      <td>-0.179525</td>\n",
              "      <td>-3.259819</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.540323</td>\n",
              "      <td>-0.318117</td>\n",
              "      <td>-0.615244</td>\n",
              "      <td>-0.875207</td>\n",
              "      <td>-3.259819</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.540323</td>\n",
              "      <td>-0.318117</td>\n",
              "      <td>-0.632003</td>\n",
              "      <td>-0.875207</td>\n",
              "      <td>-3.259819</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    anumber   bnumber  duration       fee      type\n",
              "0  1.097724 -0.319205  1.697394  0.358047 -3.259819\n",
              "1  1.097724 -0.319205  0.250598 -0.376284 -3.259819\n",
              "2  1.097724 -0.319205  0.312045 -0.179525 -3.259819\n",
              "3  0.540323 -0.318117 -0.615244 -0.875207 -3.259819\n",
              "4  0.540323 -0.318117 -0.632003 -0.875207 -3.259819"
            ]
          },
          "metadata": {},
          "execution_count": 14
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 17.8 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uJnlos1T-IXO",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "215dbec1-deb0-4941-b76a-cd1378d30604"
      },
      "source": [
        "# Show Final Dataset Description (Focus at Dtype)\n",
        "Final_Dataset.info()"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 11418 entries, 0 to 11417\n",
            "Data columns (total 5 columns):\n",
            " #   Column    Non-Null Count  Dtype  \n",
            "---  ------    --------------  -----  \n",
            " 0   anumber   11418 non-null  float64\n",
            " 1   bnumber   11418 non-null  float64\n",
            " 2   duration  11418 non-null  float64\n",
            " 3   fee       11418 non-null  float64\n",
            " 4   type      11418 non-null  float64\n",
            "dtypes: float64(5)\n",
            "memory usage: 446.1 KB\n",
            "time: 10.1 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jz-e6ybYUdjz",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        },
        "outputId": "725ca433-8fa6-4ec3-c8b4-c182e825c093"
      },
      "source": [
        "# Displays Final_Dataset Statistics (Focus at std)\n",
        "Final_Dataset.describe()"
      ],
      "execution_count": 16,
      "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>anumber</th>\n",
              "      <th>bnumber</th>\n",
              "      <th>duration</th>\n",
              "      <th>fee</th>\n",
              "      <th>type</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>1.141800e+04</td>\n",
              "      <td>1.141800e+04</td>\n",
              "      <td>1.141800e+04</td>\n",
              "      <td>1.141800e+04</td>\n",
              "      <td>1.141800e+04</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>-2.250511e-15</td>\n",
              "      <td>1.410651e-14</td>\n",
              "      <td>-6.163206e-16</td>\n",
              "      <td>-7.016049e-15</td>\n",
              "      <td>3.629611e-14</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>1.000044e+00</td>\n",
              "      <td>1.000044e+00</td>\n",
              "      <td>1.000044e+00</td>\n",
              "      <td>1.000044e+00</td>\n",
              "      <td>1.000044e+00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>-2.595056e+00</td>\n",
              "      <td>-1.770219e+00</td>\n",
              "      <td>-7.102079e-01</td>\n",
              "      <td>-1.212507e+00</td>\n",
              "      <td>-3.259819e+00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>-5.048034e-01</td>\n",
              "      <td>-9.353144e-01</td>\n",
              "      <td>-5.761419e-01</td>\n",
              "      <td>-9.314236e-01</td>\n",
              "      <td>-1.248220e+00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>-2.121680e-01</td>\n",
              "      <td>8.137480e-02</td>\n",
              "      <td>-3.918012e-01</td>\n",
              "      <td>-2.849318e-01</td>\n",
              "      <td>7.633788e-01</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>1.055919e+00</td>\n",
              "      <td>8.027996e-01</td>\n",
              "      <td>8.860188e-02</td>\n",
              "      <td>7.831852e-01</td>\n",
              "      <td>7.633788e-01</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>1.515775e+00</td>\n",
              "      <td>1.755538e+00</td>\n",
              "      <td>4.814428e+00</td>\n",
              "      <td>2.736715e+00</td>\n",
              "      <td>1.769178e+00</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "            anumber       bnumber      duration           fee          type\n",
              "count  1.141800e+04  1.141800e+04  1.141800e+04  1.141800e+04  1.141800e+04\n",
              "mean  -2.250511e-15  1.410651e-14 -6.163206e-16 -7.016049e-15  3.629611e-14\n",
              "std    1.000044e+00  1.000044e+00  1.000044e+00  1.000044e+00  1.000044e+00\n",
              "min   -2.595056e+00 -1.770219e+00 -7.102079e-01 -1.212507e+00 -3.259819e+00\n",
              "25%   -5.048034e-01 -9.353144e-01 -5.761419e-01 -9.314236e-01 -1.248220e+00\n",
              "50%   -2.121680e-01  8.137480e-02 -3.918012e-01 -2.849318e-01  7.633788e-01\n",
              "75%    1.055919e+00  8.027996e-01  8.860188e-02  7.831852e-01  7.633788e-01\n",
              "max    1.515775e+00  1.755538e+00  4.814428e+00  2.736715e+00  1.769178e+00"
            ]
          },
          "metadata": {},
          "execution_count": 16
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 30.6 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CvTNvUuTxSK-",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3e319596-dfb6-4518-e1e3-234c7fee769c"
      },
      "source": [
        "# Persiapan Dataset\n",
        "X = Final_Dataset.to_numpy()\n",
        "y = df.iloc[:, 8] # Label Fraud\n",
        "y = np.array(y).astype(np.float)"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 2.72 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6f2UMB4_ARl6",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "54167ee8-def6-4d88-f7d6-125526192853"
      },
      "source": [
        "X"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[ 1.09772399, -0.31920511,  1.6973936 ,  0.35804655, -3.25981941],\n",
              "       [ 1.09772399, -0.31920511,  0.25059827, -0.37628386, -3.25981941],\n",
              "       [ 1.09772399, -0.31920511,  0.31204517, -0.17952547, -3.25981941],\n",
              "       ...,\n",
              "       [ 1.09772399, -0.3322675 ,  0.64162403,  1.18372906,  1.76917832],\n",
              "       [ 1.09772399, -0.33117897, -0.69344965, -0.05303794,  1.76917832],\n",
              "       [ 1.09772399, -0.33117897, -0.67110532, -0.05303794,  1.76917832]])"
            ]
          },
          "metadata": {},
          "execution_count": 18
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 4.38 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "T3KEMFflAVkb",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "e5125f65-fb4b-4dca-9699-0050a80ca6af"
      },
      "source": [
        "y"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0., 0., 0., ..., 0., 0., 0.])"
            ]
          },
          "metadata": {},
          "execution_count": 19
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 3.44 ms (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1L3PMpPzG7U5"
      },
      "source": [
        "**4. DATA MODELING WITH K-MEANS**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VJhOOGy0I1gB"
      },
      "source": [
        "Kmeans algorithm is an iterative algorithm that tries to partition the dataset into Kpre-defined distinct non-overlapping subgroups (clusters) where each data point belongs to only one group. It tries to make the intra-cluster data points as similar as possible while also keeping the clusters as different (far) as possible. It assigns data points to a cluster such that the sum of the squared distance between the data points and the cluster’s centroid (arithmetic mean of all the data points that belong to that cluster) is at the minimum. The less variation we have within clusters, the more homogeneous (similar) the data points are within the same cluster. (Towards Data Science, 2018)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ck9wjGP1JDKn"
      },
      "source": [
        "**Elbow Methode**\n",
        "\n",
        "Elbow method gives us an idea on what a good k number of clusters would be based on the sum of squared distance (SSE) between data points and their assigned clusters’ centroids. We pick k at the spot where SSE starts to flatten out and forming an elbow. We’ll use the geyser dataset and evaluate SSE for different values of k and see where the curve might form an elbow and flatten out. (Towards Data Science, 2018)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OmC-jXelHv6q",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 405
        },
        "outputId": "4a253bff-c58c-48e8-c9ad-0c05248a2c52"
      },
      "source": [
        "# Find cluster with Elbow Method\n",
        "distortions = []\n",
        "K = range(1,10)\n",
        "for k in K:\n",
        "    kmeanModel = KMeans(n_clusters=k, random_state=0)\n",
        "    kmeanModel.fit(Final_Dataset)\n",
        "    distortions.append(kmeanModel.inertia_)\n",
        "\n",
        "plt.figure(figsize=(8,6))\n",
        "plt.plot(K, distortions, 'bx-')\n",
        "plt.xlabel('k')\n",
        "plt.ylabel('Distortion')\n",
        "plt.grid(c='black')\n",
        "#plt.title('The Elbow Method showing the optimal k')\n",
        "plt.show()"
      ],
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 3.2 s (started: 2021-10-11 14:32:20 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LFOp4PYtyq8o"
      },
      "source": [
        "Based on the graph above, we take the K value is 4."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jeqfsqph0aea",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8be4025d-ee05-48f2-b694-796d9ac56bc9"
      },
      "source": [
        "# K-Means parameters\n",
        "kmeans = KMeans(n_clusters=4, init ='k-means++', max_iter =1000,\n",
        "                random_state=0, verbose=0, n_init=1).fit(X)\n",
        "\n",
        "# Get the cluster number for each data point\n",
        "X_clusters = kmeans.predict(X)\n",
        "\n",
        "# Save the cluster centroid\n",
        "X_clusters_centers = kmeans.cluster_centers_\n",
        "\n",
        "# Calculate the distance to the center of the group for each point\n",
        "dist = [np.linalg.norm(x-y) for x,y in zip(X, X_clusters_centers[X_clusters])]\n",
        "\n",
        "# Make predictions based on distance\n",
        "km_y_pred = np.array(dist)\n",
        "\n",
        "# If the distance is far, it will be an outlier / fraud\n",
        "km_y_pred[dist>=np.percentile(dist, 93)] = 1\n",
        "\n",
        "# If the distance is close, it is not an outlier / non fraud\n",
        "km_y_pred[dist<np.percentile(dist, 93)] = 0\n",
        "\n",
        "km_y_pred"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([1., 1., 1., ..., 1., 0., 0.])"
            ]
          },
          "metadata": {},
          "execution_count": 21
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 197 ms (started: 2021-10-11 14:32:23 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZbhLynGj3kmB",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c314736c-e76b-4e07-bd84-f94eaa3b90bf"
      },
      "source": [
        "# Duplicate Final Dataset and Add Column Labels K-Means\n",
        "Final_Dataset_copy = Final_Dataset.copy()\n",
        "Final_Dataset_copy['labels_kmeans'] = km_y_pred"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 3.17 ms (started: 2021-10-11 14:32:23 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HusFJ2xTY1he",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 220
        },
        "outputId": "e1de8581-bd0a-49db-9e5d-7dd2433f7ec4"
      },
      "source": [
        "# Show Final Dataset Copy plus Labels K-Means \n",
        "Final_Dataset_copy.head()"
      ],
      "execution_count": 23,
      "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>anumber</th>\n",
              "      <th>bnumber</th>\n",
              "      <th>duration</th>\n",
              "      <th>fee</th>\n",
              "      <th>type</th>\n",
              "      <th>labels_kmeans</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1.097724</td>\n",
              "      <td>-0.319205</td>\n",
              "      <td>1.697394</td>\n",
              "      <td>0.358047</td>\n",
              "      <td>-3.259819</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1.097724</td>\n",
              "      <td>-0.319205</td>\n",
              "      <td>0.250598</td>\n",
              "      <td>-0.376284</td>\n",
              "      <td>-3.259819</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1.097724</td>\n",
              "      <td>-0.319205</td>\n",
              "      <td>0.312045</td>\n",
              "      <td>-0.179525</td>\n",
              "      <td>-3.259819</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.540323</td>\n",
              "      <td>-0.318117</td>\n",
              "      <td>-0.615244</td>\n",
              "      <td>-0.875207</td>\n",
              "      <td>-3.259819</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.540323</td>\n",
              "      <td>-0.318117</td>\n",
              "      <td>-0.632003</td>\n",
              "      <td>-0.875207</td>\n",
              "      <td>-3.259819</td>\n",
              "      <td>1.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    anumber   bnumber  duration       fee      type  labels_kmeans\n",
              "0  1.097724 -0.319205  1.697394  0.358047 -3.259819            1.0\n",
              "1  1.097724 -0.319205  0.250598 -0.376284 -3.259819            1.0\n",
              "2  1.097724 -0.319205  0.312045 -0.179525 -3.259819            1.0\n",
              "3  0.540323 -0.318117 -0.615244 -0.875207 -3.259819            1.0\n",
              "4  0.540323 -0.318117 -0.632003 -0.875207 -3.259819            1.0"
            ]
          },
          "metadata": {},
          "execution_count": 23
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 22.3 ms (started: 2021-10-11 14:32:23 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XajQXuQbZdgl",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "b30a062c-51c1-4c41-fcd7-f732488e92f7"
      },
      "source": [
        "# Count Labels\n",
        "Final_Dataset_copy['labels_kmeans'].value_counts().sort_index()"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.0    10618\n",
              "1.0      800\n",
              "Name: labels_kmeans, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 24
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 7.32 ms (started: 2021-10-11 14:32:23 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sZc4v4gAZ1OB"
      },
      "source": [
        "**5. DATA VISUALIZATION WITH PCA**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZAcuh1edQdPk",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "865365f2-d7c6-4f5c-f139-27e1ba1690dd"
      },
      "source": [
        "plotX = Final_Dataset_copy"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 1.18 ms (started: 2021-10-11 14:32:23 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "T4bQDj3iIMtF",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "061a6c8e-76ff-4c2b-debb-f133b4f6f2b3"
      },
      "source": [
        "pca_2d = PCA(n_components=2)"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 1.18 ms (started: 2021-10-11 14:32:24 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "drcxY-clNQq_",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6a4ccab5-480e-4695-adb1-fe447be40701"
      },
      "source": [
        "PCs_2d = pd.DataFrame(pca_2d.fit_transform(plotX.drop([\"labels_kmeans\"], axis=1)))"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 43.5 ms (started: 2021-10-11 14:32:24 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "c8E0ylMXUuhh",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7802a14f-4f5f-4de8-e89a-43afa9f006e4"
      },
      "source": [
        "PCs_2d.columns = [\"PC1_2d\", \"PC2_2d\"]"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 3.19 ms (started: 2021-10-11 14:32:24 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NdwW6ukhXesb",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8da82403-8a03-4549-a5f2-b60571c5c71a"
      },
      "source": [
        "plotX = pd.concat([plotX,PCs_2d], axis=1, join='inner')"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 4.42 ms (started: 2021-10-11 14:32:24 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oMmR7ICaZGwj",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2d66c852-0dd5-4e2c-c8c6-01683096ad4f"
      },
      "source": [
        "cluster0 = plotX[plotX[\"labels_kmeans\"] == 0]\n",
        "cluster1 = plotX[plotX[\"labels_kmeans\"] == 1]"
      ],
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 13.1 ms (started: 2021-10-11 14:32:24 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "19EJz-XmY32T",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "22720e96-e9a4-4c48-b2ff-08b318cebc2a"
      },
      "source": [
        "init_notebook_mode(connected=True)"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "        <script type=\"text/javascript\">\n",
              "        window.PlotlyConfig = {MathJaxConfig: 'local'};\n",
              "        if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}\n",
              "        if (typeof require !== 'undefined') {\n",
              "        require.undef(\"plotly\");\n",
              "        requirejs.config({\n",
              "            paths: {\n",
              "                'plotly': ['https://cdn.plot.ly/plotly-latest.min']\n",
              "            }\n",
              "        });\n",
              "        require(['plotly'], function(Plotly) {\n",
              "            window._Plotly = Plotly;\n",
              "        });\n",
              "        }\n",
              "        </script>\n",
              "        "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 4.53 ms (started: 2021-10-11 14:32:24 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kX3akTE2Zd1p",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "cd29227a-e3c5-42cd-d6df-15e21c9f0749"
      },
      "source": [
        "#trace1 is for 'Cluster 0'\n",
        "trace1 = go.Scatter(\n",
        "                    x = cluster0[\"PC1_2d\"],\n",
        "                    y = cluster0[\"PC2_2d\"],\n",
        "                    mode = \"markers\",\n",
        "                    name = \"Non Fraud\",\n",
        "                    marker = dict(color = 'rgba(0, 0, 0, 0.8)'),\n",
        "                    text = None)\n",
        "\n",
        "#trace2 is for 'Cluster 1'\n",
        "trace2 = go.Scatter(\n",
        "                    x = cluster1[\"PC1_2d\"],\n",
        "                    y = cluster1[\"PC2_2d\"],\n",
        "                    mode = \"markers\",\n",
        "                    name = \"Fraud\",\n",
        "                    marker = dict(color = 'rgba(240, 52, 52, 0.8)'),\n",
        "                    text = None)\n",
        "\n",
        "data = [trace1, trace2]\n",
        "\n",
        "title = \"K-Means - PCA\"\n",
        "\n",
        "layout = dict(title = title,\n",
        "              xaxis= dict(title= 'PC1',ticklen= 5,zeroline= False),\n",
        "              yaxis= dict(title= 'PC2',ticklen= 5,zeroline= False)\n",
        "             )\n",
        "\n",
        "#Plot and save html\n",
        "plotly.offline.plot({\"data\": data,\n",
        "                     \"layout\": layout},\n",
        "                     auto_open=True)"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'temp-plot.html'"
            ]
          },
          "metadata": {},
          "execution_count": 32
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 1.24 s (started: 2021-10-11 14:32:24 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0BvOTvWMasU4"
      },
      "source": [
        "Download and Open File temp-plot.html to see the Graph, please screenshoot and Place in here.\n",
        "\n",
        "![](https://github.com/multimediary/portfolio-data-science/blob/main/screenshots/fraud-modeling-CDR-K-Means.jpg?raw=true)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GHRJdugewQgj"
      },
      "source": [
        "**6. DATA EVALUATION WITH CONFUSION MATRIX**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "07SkEFCTmj3a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 301
        },
        "outputId": "671cc117-8975-4844-87ed-71721d5e79a5"
      },
      "source": [
        "C = confusion_matrix(df[\"label\"],plotX[\"labels_kmeans\"])\n",
        "\n",
        "# Plot confusion matrix\n",
        "plt.imshow(C,interpolation='none',cmap='Blues')\n",
        "for (i, j), z in np.ndenumerate(C):\n",
        "    plt.text(j, i, z, ha='center', va='center')\n",
        "plt.xlabel(\"labels_kmeans\")\n",
        "plt.ylabel(\"label_fraud\")\n",
        "plt.show()"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "time: 191 ms (started: 2021-10-11 14:32:25 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7Ka9UgbCm4rh",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "32ea0f8b-936a-4863-875e-3f36d222cd50"
      },
      "source": [
        "print(C)"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[10238   783]\n",
            " [  380    17]]\n",
            "time: 2.11 ms (started: 2021-10-11 14:32:25 +00:00)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fdm51Dbom72l",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f9755b0f-66a3-496e-f585-f2c96d871254"
      },
      "source": [
        "# Nilai Accuracy, Precision, Recall, and F1 Score\n",
        "print(\"Accuracy Score of K-Means\", accuracy_score(df[\"label\"], plotX[\"labels_kmeans\"]))\n",
        "print(\"Recall of K-Means\", recall_score(df[\"label\"], plotX[\"labels_kmeans\"],average=None, zero_division=1))\n",
        "print(\"Precision Score of K-Means\", precision_score(df[\"label\"], plotX[\"labels_kmeans\"],average=None, zero_division=1))\n",
        "print(\"F1 Score of K-Means\", f1_score(df[\"label\"], plotX[\"labels_kmeans\"],average=None, zero_division=1))"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Accuracy Score of K-Means 0.8981432825363461\n",
            "Recall of K-Means [0.92895382 0.04282116]\n",
            "Precision Score of K-Means [0.96421172 0.02125   ]\n",
            "F1 Score of K-Means [0.94625445 0.02840434]\n",
            "time: 35.6 ms (started: 2021-10-11 14:32:25 +00:00)\n"
          ]
        }
      ]
    }
  ]
}