Project_DTI_Machine_Learning_Method_MLP_Method.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Project DTI Machine Learning Method MLP Method",
      "provenance": [],
      "collapsed_sections": [
        "oxnAOntXv2Sf",
        "yTBhSYpxv5Xu",
        "e1eHsSQX0bbG",
        "SM31GdcSzhif",
        "_GCRrnL6yxNF",
        "8NgEsPaPy0fk",
        "adKQpmsUy4yU",
        "wTb2gXzTwP_6",
        "sD6oXyPqy7HW",
        "CusfZQ-SkIVn",
        "YnF_WVM8svOq"
      ],
      "include_colab_link": true
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/alfhi24/FinalProjectDTI/blob/main/Project_DTI_Machine_Learning_Method_MLP_Method.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "oxnAOntXv2Sf"
      },
      "source": [
        "### **Import Library**\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mHNEIi8JJVdR",
        "outputId": "f90a976d-441a-4d52-fadd-fd1989e09546"
      },
      "source": [
        "! pip3 install Sastrawi"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting Sastrawi\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/6f/4b/bab676953da3103003730b8fcdfadbdd20f333d4add10af949dd5c51e6ed/Sastrawi-1.0.1-py2.py3-none-any.whl (209kB)\n",
            "\r\u001b[K     |█▋                              | 10kB 15.4MB/s eta 0:00:01\r\u001b[K     |███▏                            | 20kB 20.3MB/s eta 0:00:01\r\u001b[K     |████▊                           | 30kB 18.2MB/s eta 0:00:01\r\u001b[K     |██████▎                         | 40kB 12.4MB/s eta 0:00:01\r\u001b[K     |███████▉                        | 51kB 7.4MB/s eta 0:00:01\r\u001b[K     |█████████▍                      | 61kB 8.0MB/s eta 0:00:01\r\u001b[K     |███████████                     | 71kB 8.9MB/s eta 0:00:01\r\u001b[K     |████████████▌                   | 81kB 8.1MB/s eta 0:00:01\r\u001b[K     |██████████████                  | 92kB 8.7MB/s eta 0:00:01\r\u001b[K     |███████████████▋                | 102kB 9.0MB/s eta 0:00:01\r\u001b[K     |█████████████████▏              | 112kB 9.0MB/s eta 0:00:01\r\u001b[K     |██████████████████▊             | 122kB 9.0MB/s eta 0:00:01\r\u001b[K     |████████████████████▎           | 133kB 9.0MB/s eta 0:00:01\r\u001b[K     |█████████████████████▉          | 143kB 9.0MB/s eta 0:00:01\r\u001b[K     |███████████████████████▍        | 153kB 9.0MB/s eta 0:00:01\r\u001b[K     |█████████████████████████       | 163kB 9.0MB/s eta 0:00:01\r\u001b[K     |██████████████████████████▋     | 174kB 9.0MB/s eta 0:00:01\r\u001b[K     |████████████████████████████▏   | 184kB 9.0MB/s eta 0:00:01\r\u001b[K     |█████████████████████████████▊  | 194kB 9.0MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████▎| 204kB 9.0MB/s eta 0:00:01\r\u001b[K     |████████████████████████████████| 215kB 9.0MB/s \n",
            "\u001b[?25hInstalling collected packages: Sastrawi\n",
            "Successfully installed Sastrawi-1.0.1\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1rEEvzvrHExs",
        "outputId": "1d0763dd-f884-4ee8-ca5b-8312787c7104"
      },
      "source": [
        "! pip3 install scikit-plot"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting scikit-plot\n",
            "  Downloading https://files.pythonhosted.org/packages/7c/47/32520e259340c140a4ad27c1b97050dd3254fdc517b1d59974d47037510e/scikit_plot-0.3.7-py3-none-any.whl\n",
            "Requirement already satisfied: scikit-learn>=0.18 in /usr/local/lib/python3.6/dist-packages (from scikit-plot) (0.22.2.post1)\n",
            "Requirement already satisfied: matplotlib>=1.4.0 in /usr/local/lib/python3.6/dist-packages (from scikit-plot) (3.2.2)\n",
            "Requirement already satisfied: scipy>=0.9 in /usr/local/lib/python3.6/dist-packages (from scikit-plot) (1.4.1)\n",
            "Requirement already satisfied: joblib>=0.10 in /usr/local/lib/python3.6/dist-packages (from scikit-plot) (0.17.0)\n",
            "Requirement already satisfied: numpy>=1.11.0 in /usr/local/lib/python3.6/dist-packages (from scikit-learn>=0.18->scikit-plot) (1.18.5)\n",
            "Requirement already satisfied: python-dateutil>=2.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=1.4.0->scikit-plot) (2.8.1)\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>=1.4.0->scikit-plot) (2.4.7)\n",
            "Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=1.4.0->scikit-plot) (0.10.0)\n",
            "Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.6/dist-packages (from matplotlib>=1.4.0->scikit-plot) (1.3.1)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.6/dist-packages (from python-dateutil>=2.1->matplotlib>=1.4.0->scikit-plot) (1.15.0)\n",
            "Installing collected packages: scikit-plot\n",
            "Successfully installed scikit-plot-0.3.7\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yOpFmlP5tT5t"
      },
      "source": [
        "# Library for dataset\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "\n",
        "# Library for data visualization\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "import plotly.express as px\n",
        "from os import path\n",
        "from PIL import Image\n",
        "from wordcloud import WordCloud, STOPWORDS, ImageColorGenerator\n",
        "import matplotlib.pyplot as plt\n",
        "import warnings\n",
        "warnings.filterwarnings(\"ignore\")\n",
        "\n",
        "# Library for preprocessing\n",
        "import re\n",
        "import nltk\n",
        "from nltk.corpus import stopwords\n",
        "from Sastrawi.Stemmer.StemmerFactory import StemmerFactory\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "from imblearn.over_sampling import SMOTE\n",
        "from imblearn.under_sampling import RandomUnderSampler\n",
        "from sklearn.model_selection import train_test_split, cross_val_score\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "\n",
        "#Library for classification\n",
        "from sklearn.model_selection import StratifiedKFold, cross_validate\n",
        "from sklearn.naive_bayes import GaussianNB \n",
        "from sklearn.svm import SVC\n",
        "from sklearn.neural_network import MLPClassifier\n",
        "from sklearn.neighbors import KNeighborsClassifier\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.model_selection import GridSearchCV\n",
        "from sklearn.model_selection import RandomizedSearchCV\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "\n",
        "#Library for Evaluation\n",
        "from sklearn import metrics\n",
        "import time\n",
        "from sklearn.metrics import classification_report, accuracy_score\n",
        "from mlxtend.plotting import plot_confusion_matrix\n",
        "from scikitplot.metrics import plot_confusion_matrix as pcm\n",
        "\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bUeutW80gf_O",
        "outputId": "f1cce6c6-bde1-4463-a4f7-8cc3ac84c2ab"
      },
      "source": [
        "nltk.download('punkt')\n",
        "nltk.download('stopwords')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[nltk_data] Downloading package punkt to /root/nltk_data...\n",
            "[nltk_data]   Unzipping tokenizers/punkt.zip.\n",
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Unzipping corpora/stopwords.zip.\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yTBhSYpxv5Xu"
      },
      "source": [
        "### **Dataset**\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Oat6ZOVqv8By",
        "outputId": "ddb3d9ea-977c-407e-cefe-b182658d56ba"
      },
      "source": [
        "df = pd.read_csv('https://raw.githubusercontent.com/asthala/racism-detection/master/datasetfix.csv')\n",
        "df.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>tweets</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Trus nih buat yg nanya kenapa gak dibangun di ...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>PLAY &amp; Let The Fun Begin 😱\\nLike + RT + Click ...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Papua adalah bagian dari NKRI\\n#DukungOtsusPap...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Akar konflik harus dibicarakan.\\nDan dicari ti...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>\"Saya minta kita semua jadi orang Papua. Artin...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                              tweets  label\n",
              "0  Trus nih buat yg nanya kenapa gak dibangun di ...  Non_R\n",
              "1  PLAY & Let The Fun Begin 😱\\nLike + RT + Click ...  Non_R\n",
              "2  Papua adalah bagian dari NKRI\\n#DukungOtsusPap...  Non_R\n",
              "3  Akar konflik harus dibicarakan.\\nDan dicari ti...  Non_R\n",
              "4  \"Saya minta kita semua jadi orang Papua. Artin...  Non_R"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GZVFLbhXxPra",
        "outputId": "a6ab7adc-e09f-4e1f-ad71-69cb5180515c"
      },
      "source": [
        "count_label =df.label.value_counts().to_frame()\n",
        "px.bar(data_frame=count_label, x=count_label.index , y='label')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "<html>\n",
              "<head><meta charset=\"utf-8\" /></head>\n",
              "<body>\n",
              "    <div>\n",
              "            <script src=\"https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js?config=TeX-AMS-MML_SVG\"></script><script type=\"text/javascript\">if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}</script>\n",
              "                <script type=\"text/javascript\">window.PlotlyConfig = {MathJaxConfig: 'local'};</script>\n",
              "        <script src=\"https://cdn.plot.ly/plotly-latest.min.js\"></script>    \n",
              "            <div id=\"b84a2184-df51-4d93-88c9-0d34ee47ce93\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>\n",
              "            <script type=\"text/javascript\">\n",
              "                \n",
              "                    window.PLOTLYENV=window.PLOTLYENV || {};\n",
              "                    \n",
              "                if (document.getElementById(\"b84a2184-df51-4d93-88c9-0d34ee47ce93\")) {\n",
              "                    Plotly.newPlot(\n",
              "                        'b84a2184-df51-4d93-88c9-0d34ee47ce93',\n",
              "                        [{\"alignmentgroup\": \"True\", \"hoverlabel\": {\"namelength\": 0}, \"hovertemplate\": \"x=%{x}<br>label=%{y}\", \"legendgroup\": \"\", \"marker\": {\"color\": \"#636efa\"}, \"name\": \"\", \"offsetgroup\": \"\", \"orientation\": \"v\", \"showlegend\": false, \"textposition\": \"auto\", \"type\": \"bar\", \"x\": [\"Non_R\", \"R\"], \"xaxis\": \"x\", \"y\": [511, 175], \"yaxis\": \"y\"}],\n",
              "                        {\"barmode\": \"relative\", \"legend\": {\"tracegroupgap\": 0}, \"margin\": {\"t\": 60}, \"template\": {\"data\": {\"bar\": [{\"error_x\": {\"color\": \"#2a3f5f\"}, \"error_y\": {\"color\": \"#2a3f5f\"}, \"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"bar\"}], \"barpolar\": [{\"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"barpolar\"}], \"carpet\": [{\"aaxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"baxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"type\": \"carpet\"}], \"choropleth\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"choropleth\"}], \"contour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"contour\"}], \"contourcarpet\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"contourcarpet\"}], \"heatmap\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmap\"}], \"heatmapgl\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmapgl\"}], \"histogram\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"histogram\"}], \"histogram2d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2d\"}], \"histogram2dcontour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2dcontour\"}], \"mesh3d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"mesh3d\"}], \"parcoords\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"parcoords\"}], \"pie\": [{\"automargin\": true, \"type\": \"pie\"}], \"scatter\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter\"}], \"scatter3d\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter3d\"}], \"scattercarpet\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattercarpet\"}], \"scattergeo\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergeo\"}], \"scattergl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergl\"}], \"scattermapbox\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattermapbox\"}], \"scatterpolar\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolar\"}], \"scatterpolargl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolargl\"}], \"scatterternary\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterternary\"}], \"surface\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"surface\"}], \"table\": [{\"cells\": {\"fill\": {\"color\": \"#EBF0F8\"}, \"line\": {\"color\": \"white\"}}, \"header\": {\"fill\": {\"color\": \"#C8D4E3\"}, \"line\": {\"color\": \"white\"}}, \"type\": \"table\"}]}, \"layout\": {\"annotationdefaults\": {\"arrowcolor\": \"#2a3f5f\", \"arrowhead\": 0, \"arrowwidth\": 1}, \"coloraxis\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"colorscale\": {\"diverging\": [[0, \"#8e0152\"], [0.1, \"#c51b7d\"], [0.2, \"#de77ae\"], [0.3, \"#f1b6da\"], [0.4, \"#fde0ef\"], [0.5, \"#f7f7f7\"], [0.6, \"#e6f5d0\"], [0.7, \"#b8e186\"], [0.8, \"#7fbc41\"], [0.9, \"#4d9221\"], [1, \"#276419\"]], \"sequential\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"sequentialminus\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]]}, \"colorway\": [\"#636efa\", \"#EF553B\", \"#00cc96\", \"#ab63fa\", \"#FFA15A\", \"#19d3f3\", \"#FF6692\", \"#B6E880\", \"#FF97FF\", \"#FECB52\"], \"font\": {\"color\": \"#2a3f5f\"}, \"geo\": {\"bgcolor\": \"white\", \"lakecolor\": \"white\", \"landcolor\": \"#E5ECF6\", \"showlakes\": true, \"showland\": true, \"subunitcolor\": \"white\"}, \"hoverlabel\": {\"align\": \"left\"}, \"hovermode\": \"closest\", \"mapbox\": {\"style\": \"light\"}, \"paper_bgcolor\": \"white\", \"plot_bgcolor\": \"#E5ECF6\", \"polar\": {\"angularaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"radialaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"scene\": {\"xaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"yaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"zaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}}, \"shapedefaults\": {\"line\": {\"color\": \"#2a3f5f\"}}, \"ternary\": {\"aaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"baxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"caxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"title\": {\"x\": 0.05}, \"xaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"title\": {\"standoff\": 15}, \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}, \"yaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"title\": {\"standoff\": 15}, \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}}}, \"xaxis\": {\"anchor\": \"y\", \"domain\": [0.0, 1.0], \"title\": {\"text\": \"x\"}}, \"yaxis\": {\"anchor\": \"x\", \"domain\": [0.0, 1.0], \"title\": {\"text\": \"label\"}}},\n",
              "                        {\"responsive\": true}\n",
              "                    ).then(function(){\n",
              "                            \n",
              "var gd = document.getElementById('b84a2184-df51-4d93-88c9-0d34ee47ce93');\n",
              "var x = new MutationObserver(function (mutations, observer) {{\n",
              "        var display = window.getComputedStyle(gd).display;\n",
              "        if (!display || display === 'none') {{\n",
              "            console.log([gd, 'removed!']);\n",
              "            Plotly.purge(gd);\n",
              "            observer.disconnect();\n",
              "        }}\n",
              "}});\n",
              "\n",
              "// Listen for the removal of the full notebook cells\n",
              "var notebookContainer = gd.closest('#notebook-container');\n",
              "if (notebookContainer) {{\n",
              "    x.observe(notebookContainer, {childList: true});\n",
              "}}\n",
              "\n",
              "// Listen for the clearing of the current output cell\n",
              "var outputEl = gd.closest('.output');\n",
              "if (outputEl) {{\n",
              "    x.observe(outputEl, {childList: true});\n",
              "}}\n",
              "\n",
              "                        })\n",
              "                };\n",
              "                \n",
              "            </script>\n",
              "        </div>\n",
              "</body>\n",
              "</html>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vZZj3oqfxjvd"
      },
      "source": [
        "Dari barchart diatas, dataset yang digunakan memiliki imbalance data label, sehingga pada tahap preprocessing akan dilakukan SMOTE untuk menanggulangi data yang tidak seimbang"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9o8xOnCPyrzH"
      },
      "source": [
        "### **Preprocessing**"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "e1eHsSQX0bbG"
      },
      "source": [
        "#### Data Cleaning"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8jpK5sgz0ei0"
      },
      "source": [
        "def clean(data):\n",
        "  tweets = []\n",
        "  for tweet in data:\n",
        "      tweet = re.sub(r\"'(?:\\@|https?\\://)\\S+\", \"\", tweet)\n",
        "      tweet = re.sub('\\n', '', tweet)\n",
        "      tweet = re.sub('rt', '', tweet)\n",
        "      tweet = re.sub(\"[^a-zA-Z^']\", \" \", tweet)\n",
        "      tweet = re.sub(\" {2,}\", \" \", tweet)\n",
        "      tweet = tweet.strip()\n",
        "      tweets.append(tweet)\n",
        "  return tweets"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "L46eh-b_n00N",
        "outputId": "72deeda5-14ed-4331-df1a-5f9e90b3bdcd"
      },
      "source": [
        "df_clean = df.copy()\n",
        "df_clean['tweets']= clean(df['tweets'])\n",
        "df_clean.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>tweets</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Trus nih buat yg nanya kenapa gak dibangun di ...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>PLAY Let The Fun Begin Like RT Click https www...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Papua adalah bagian dari NKRI DukungOtsusPapua...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Akar konflik harus dibicarakan Dan dicari titi...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>Saya minta kita semua jadi orang Papua Ainya k...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                              tweets  label\n",
              "0  Trus nih buat yg nanya kenapa gak dibangun di ...  Non_R\n",
              "1  PLAY Let The Fun Begin Like RT Click https www...  Non_R\n",
              "2  Papua adalah bagian dari NKRI DukungOtsusPapua...  Non_R\n",
              "3  Akar konflik harus dibicarakan Dan dicari titi...  Non_R\n",
              "4  Saya minta kita semua jadi orang Papua Ainya k...  Non_R"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SM31GdcSzhif"
      },
      "source": [
        "#### Case Folding\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yk-ECjYf0PNd"
      },
      "source": [
        "def case_fold(data):\n",
        "    return data.str.lower()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "llNDHgXpntHP",
        "outputId": "2a19459a-81b5-4f17-81e1-fb7027520648"
      },
      "source": [
        "df_clean['tweets'] = case_fold(df_clean['tweets'])\n",
        "df_clean.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>tweets</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>trus nih buat yg nanya kenapa gak dibangun di ...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>play let the fun begin like rt click https www...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>papua adalah bagian dari nkri dukungotsuspapua...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>akar konflik harus dibicarakan dan dicari titi...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>saya minta kita semua jadi orang papua ainya k...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                              tweets  label\n",
              "0  trus nih buat yg nanya kenapa gak dibangun di ...  Non_R\n",
              "1  play let the fun begin like rt click https www...  Non_R\n",
              "2  papua adalah bagian dari nkri dukungotsuspapua...  Non_R\n",
              "3  akar konflik harus dibicarakan dan dicari titi...  Non_R\n",
              "4  saya minta kita semua jadi orang papua ainya k...  Non_R"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_GCRrnL6yxNF"
      },
      "source": [
        "#### Tokenization\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sjzQrdVKoE79"
      },
      "source": [
        "def token(data):\n",
        "  return data.apply(nltk.word_tokenize)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zCXhQO9EwzF9",
        "outputId": "0d80e167-cfe3-4f29-afde-2a929879397c"
      },
      "source": [
        "df_clean['tweets'] = token(df_clean['tweets'])\n",
        "df_clean.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>tweets</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>[trus, nih, buat, yg, nanya, kenapa, gak, diba...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>[play, let, the, fun, begin, like, rt, click, ...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>[papua, adalah, bagian, dari, nkri, dukungotsu...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>[akar, konflik, harus, dibicarakan, dan, dicar...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>[saya, minta, kita, semua, jadi, orang, papua,...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                              tweets  label\n",
              "0  [trus, nih, buat, yg, nanya, kenapa, gak, diba...  Non_R\n",
              "1  [play, let, the, fun, begin, like, rt, click, ...  Non_R\n",
              "2  [papua, adalah, bagian, dari, nkri, dukungotsu...  Non_R\n",
              "3  [akar, konflik, harus, dibicarakan, dan, dicar...  Non_R\n",
              "4  [saya, minta, kita, semua, jadi, orang, papua,...  Non_R"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8NgEsPaPy0fk"
      },
      "source": [
        "#### Stopword\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LfpCMxsPrTWL"
      },
      "source": [
        "def stop_words(data) :\n",
        "  stop_words = set(stopwords.words('indonesian'))\n",
        "  return data.apply(lambda x: [item for item in x if item not in stop_words])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "98gyvnPTy4oD",
        "outputId": "68286f28-89e6-4a29-a9bb-ab4127f58e4c"
      },
      "source": [
        "df_clean['tweets'] = stop_words(df_clean['tweets'])\n",
        "df_clean.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>tweets</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>[trus, nih, yg, nanya, gak, dibangun, papua, a...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>[play, let, the, fun, begin, like, rt, click, ...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>[papua, nkri, dukungotsuspapuamaju]</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>[akar, konflik, dibicarakan, dicari, titik, te...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>[orang, papua, ainya, kulit, hitam, kulit, put...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                              tweets  label\n",
              "0  [trus, nih, yg, nanya, gak, dibangun, papua, a...  Non_R\n",
              "1  [play, let, the, fun, begin, like, rt, click, ...  Non_R\n",
              "2                [papua, nkri, dukungotsuspapuamaju]  Non_R\n",
              "3  [akar, konflik, dibicarakan, dicari, titik, te...  Non_R\n",
              "4  [orang, papua, ainya, kulit, hitam, kulit, put...  Non_R"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "adKQpmsUy4yU"
      },
      "source": [
        "#### Stemming\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JFKWV1eSs1QB"
      },
      "source": [
        "def stem(data):\n",
        "  factory = StemmerFactory()\n",
        "  stemmer = factory.create_stemmer()\n",
        "  return data.apply(lambda x: [ stemmer.stem(item) for item in x])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QgI4BMFMy6go",
        "outputId": "17a0246a-3124-4604-e297-5b831a59dc21"
      },
      "source": [
        "df_clean['tweets'] = stem(df_clean['tweets'])\n",
        "df_clean.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>tweets</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>[trus, nih, yg, nanya, gak, bangun, papua, aja...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>[play, let, the, fun, begin, like, rt, click, ...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>[papua, nkri, dukungotsuspapuamaju]</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>[akar, konflik, bicara, cari, titik, temu, sel...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>[orang, papua, ainya, kulit, hitam, kulit, put...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                              tweets  label\n",
              "0  [trus, nih, yg, nanya, gak, bangun, papua, aja...  Non_R\n",
              "1  [play, let, the, fun, begin, like, rt, click, ...  Non_R\n",
              "2                [papua, nkri, dukungotsuspapuamaju]  Non_R\n",
              "3  [akar, konflik, bicara, cari, titik, temu, sel...  Non_R\n",
              "4  [orang, papua, ainya, kulit, hitam, kulit, put...  Non_R"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wTb2gXzTwP_6"
      },
      "source": [
        "#### Racism Tweet Word Cloud"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2F0B2vaeqU6M",
        "outputId": "1c97dfd2-31b5-4276-c317-2ef4e929aeb6"
      },
      "source": [
        "# Create and generate a word cloud image:\n",
        "word = df_clean.tweets[df_clean['label']=='R'].astype('string')\n",
        "word.head()\n",
        "wordcloud = WordCloud(max_font_size=50, max_words=100, background_color=\"white\").generate_from_text(' '.join(word))\n",
        "\n",
        "# Display the generated image:\n",
        "plt.imshow(wordcloud, interpolation='bilinear')\n",
        "plt.axis(\"off\")\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": "markdown",
      "metadata": {
        "id": "SQuScH-xzDh_"
      },
      "source": [
        "#### TF-IDF\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CR_qz-zvuvKR"
      },
      "source": [
        "def tf_idf(data):\n",
        "  vectorizer = TfidfVectorizer()\n",
        "  return vectorizer, vectorizer.fit_transform(data)\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 194
        },
        "id": "Ewbq5bAIpek2",
        "outputId": "43b8fc15-db69-43db-f785-0fa4eeb098aa"
      },
      "source": [
        "df_clean['tweets'] = df_clean['tweets'].apply(lambda x: \" \".join(x) if isinstance(x, list) else x)\n",
        "df_clean.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>tweets</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>trus nih yg nanya gak bangun papua aja provins...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>play let the fun begin like rt click https www...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>papua nkri dukungotsuspapuamaju</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>akar konflik bicara cari titik temu selesai sy...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>orang papua ainya kulit hitam kulit putih ramb...</td>\n",
              "      <td>Non_R</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                              tweets  label\n",
              "0  trus nih yg nanya gak bangun papua aja provins...  Non_R\n",
              "1  play let the fun begin like rt click https www...  Non_R\n",
              "2                    papua nkri dukungotsuspapuamaju  Non_R\n",
              "3  akar konflik bicara cari titik temu selesai sy...  Non_R\n",
              "4  orang papua ainya kulit hitam kulit putih ramb...  Non_R"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 36
        },
        "id": "dBm3BOkyto19",
        "outputId": "ca6fb26a-b559-456f-d73a-84c8fa99a33c"
      },
      "source": [
        "df_clean['tweets'][0]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'trus nih yg nanya gak bangun papua aja provinsi timur indonesia milik infrastruktur fasilitas olah lanjut pada freepo'"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "V8TzJSkjzGPh"
      },
      "source": [
        "vec,x = tf_idf(df_clean['tweets'])\n",
        "label = np.array(df_clean['label'].values)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sD6oXyPqy7HW"
      },
      "source": [
        "#### SMOTE"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mCmHMA9vy9qe"
      },
      "source": [
        "sampler = SMOTE()\n",
        "x_so, label_so = sampler.fit_resample(x,label)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Owr7fa3Hvswi"
      },
      "source": [
        "Label_so = pd.DataFrame(data=label_so, columns=['label'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 542
        },
        "id": "EYa19-5NvhXC",
        "outputId": "c16ab9e7-83f1-482a-e860-2512ad12088f"
      },
      "source": [
        "count_label = Label_so['label'].value_counts().to_frame()\n",
        "px.bar(data_frame=count_label, x=count_label.index , y='label')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/html": [
              "<html>\n",
              "<head><meta charset=\"utf-8\" /></head>\n",
              "<body>\n",
              "    <div>\n",
              "            <script src=\"https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js?config=TeX-AMS-MML_SVG\"></script><script type=\"text/javascript\">if (window.MathJax) {MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}</script>\n",
              "                <script type=\"text/javascript\">window.PlotlyConfig = {MathJaxConfig: 'local'};</script>\n",
              "        <script src=\"https://cdn.plot.ly/plotly-latest.min.js\"></script>    \n",
              "            <div id=\"ab76c1e9-4e43-441a-944d-f5ccf129b8e3\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div>\n",
              "            <script type=\"text/javascript\">\n",
              "                \n",
              "                    window.PLOTLYENV=window.PLOTLYENV || {};\n",
              "                    \n",
              "                if (document.getElementById(\"ab76c1e9-4e43-441a-944d-f5ccf129b8e3\")) {\n",
              "                    Plotly.newPlot(\n",
              "                        'ab76c1e9-4e43-441a-944d-f5ccf129b8e3',\n",
              "                        [{\"alignmentgroup\": \"True\", \"hoverlabel\": {\"namelength\": 0}, \"hovertemplate\": \"x=%{x}<br>label=%{y}\", \"legendgroup\": \"\", \"marker\": {\"color\": \"#636efa\"}, \"name\": \"\", \"offsetgroup\": \"\", \"orientation\": \"v\", \"showlegend\": false, \"textposition\": \"auto\", \"type\": \"bar\", \"x\": [\"R\", \"Non_R\"], \"xaxis\": \"x\", \"y\": [511, 511], \"yaxis\": \"y\"}],\n",
              "                        {\"barmode\": \"relative\", \"legend\": {\"tracegroupgap\": 0}, \"margin\": {\"t\": 60}, \"template\": {\"data\": {\"bar\": [{\"error_x\": {\"color\": \"#2a3f5f\"}, \"error_y\": {\"color\": \"#2a3f5f\"}, \"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"bar\"}], \"barpolar\": [{\"marker\": {\"line\": {\"color\": \"#E5ECF6\", \"width\": 0.5}}, \"type\": \"barpolar\"}], \"carpet\": [{\"aaxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"baxis\": {\"endlinecolor\": \"#2a3f5f\", \"gridcolor\": \"white\", \"linecolor\": \"white\", \"minorgridcolor\": \"white\", \"startlinecolor\": \"#2a3f5f\"}, \"type\": \"carpet\"}], \"choropleth\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"choropleth\"}], \"contour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"contour\"}], \"contourcarpet\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"contourcarpet\"}], \"heatmap\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmap\"}], \"heatmapgl\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"heatmapgl\"}], \"histogram\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"histogram\"}], \"histogram2d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2d\"}], \"histogram2dcontour\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"histogram2dcontour\"}], \"mesh3d\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"type\": \"mesh3d\"}], \"parcoords\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"parcoords\"}], \"pie\": [{\"automargin\": true, \"type\": \"pie\"}], \"scatter\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter\"}], \"scatter3d\": [{\"line\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatter3d\"}], \"scattercarpet\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattercarpet\"}], \"scattergeo\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergeo\"}], \"scattergl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattergl\"}], \"scattermapbox\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scattermapbox\"}], \"scatterpolar\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolar\"}], \"scatterpolargl\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterpolargl\"}], \"scatterternary\": [{\"marker\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"type\": \"scatterternary\"}], \"surface\": [{\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}, \"colorscale\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"type\": \"surface\"}], \"table\": [{\"cells\": {\"fill\": {\"color\": \"#EBF0F8\"}, \"line\": {\"color\": \"white\"}}, \"header\": {\"fill\": {\"color\": \"#C8D4E3\"}, \"line\": {\"color\": \"white\"}}, \"type\": \"table\"}]}, \"layout\": {\"annotationdefaults\": {\"arrowcolor\": \"#2a3f5f\", \"arrowhead\": 0, \"arrowwidth\": 1}, \"coloraxis\": {\"colorbar\": {\"outlinewidth\": 0, \"ticks\": \"\"}}, \"colorscale\": {\"diverging\": [[0, \"#8e0152\"], [0.1, \"#c51b7d\"], [0.2, \"#de77ae\"], [0.3, \"#f1b6da\"], [0.4, \"#fde0ef\"], [0.5, \"#f7f7f7\"], [0.6, \"#e6f5d0\"], [0.7, \"#b8e186\"], [0.8, \"#7fbc41\"], [0.9, \"#4d9221\"], [1, \"#276419\"]], \"sequential\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]], \"sequentialminus\": [[0.0, \"#0d0887\"], [0.1111111111111111, \"#46039f\"], [0.2222222222222222, \"#7201a8\"], [0.3333333333333333, \"#9c179e\"], [0.4444444444444444, \"#bd3786\"], [0.5555555555555556, \"#d8576b\"], [0.6666666666666666, \"#ed7953\"], [0.7777777777777778, \"#fb9f3a\"], [0.8888888888888888, \"#fdca26\"], [1.0, \"#f0f921\"]]}, \"colorway\": [\"#636efa\", \"#EF553B\", \"#00cc96\", \"#ab63fa\", \"#FFA15A\", \"#19d3f3\", \"#FF6692\", \"#B6E880\", \"#FF97FF\", \"#FECB52\"], \"font\": {\"color\": \"#2a3f5f\"}, \"geo\": {\"bgcolor\": \"white\", \"lakecolor\": \"white\", \"landcolor\": \"#E5ECF6\", \"showlakes\": true, \"showland\": true, \"subunitcolor\": \"white\"}, \"hoverlabel\": {\"align\": \"left\"}, \"hovermode\": \"closest\", \"mapbox\": {\"style\": \"light\"}, \"paper_bgcolor\": \"white\", \"plot_bgcolor\": \"#E5ECF6\", \"polar\": {\"angularaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"radialaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"scene\": {\"xaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"yaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}, \"zaxis\": {\"backgroundcolor\": \"#E5ECF6\", \"gridcolor\": \"white\", \"gridwidth\": 2, \"linecolor\": \"white\", \"showbackground\": true, \"ticks\": \"\", \"zerolinecolor\": \"white\"}}, \"shapedefaults\": {\"line\": {\"color\": \"#2a3f5f\"}}, \"ternary\": {\"aaxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"baxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}, \"bgcolor\": \"#E5ECF6\", \"caxis\": {\"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\"}}, \"title\": {\"x\": 0.05}, \"xaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"title\": {\"standoff\": 15}, \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}, \"yaxis\": {\"automargin\": true, \"gridcolor\": \"white\", \"linecolor\": \"white\", \"ticks\": \"\", \"title\": {\"standoff\": 15}, \"zerolinecolor\": \"white\", \"zerolinewidth\": 2}}}, \"xaxis\": {\"anchor\": \"y\", \"domain\": [0.0, 1.0], \"title\": {\"text\": \"x\"}}, \"yaxis\": {\"anchor\": \"x\", \"domain\": [0.0, 1.0], \"title\": {\"text\": \"label\"}}},\n",
              "                        {\"responsive\": true}\n",
              "                    ).then(function(){\n",
              "                            \n",
              "var gd = document.getElementById('ab76c1e9-4e43-441a-944d-f5ccf129b8e3');\n",
              "var x = new MutationObserver(function (mutations, observer) {{\n",
              "        var display = window.getComputedStyle(gd).display;\n",
              "        if (!display || display === 'none') {{\n",
              "            console.log([gd, 'removed!']);\n",
              "            Plotly.purge(gd);\n",
              "            observer.disconnect();\n",
              "        }}\n",
              "}});\n",
              "\n",
              "// Listen for the removal of the full notebook cells\n",
              "var notebookContainer = gd.closest('#notebook-container');\n",
              "if (notebookContainer) {{\n",
              "    x.observe(notebookContainer, {childList: true});\n",
              "}}\n",
              "\n",
              "// Listen for the clearing of the current output cell\n",
              "var outputEl = gd.closest('.output');\n",
              "if (outputEl) {{\n",
              "    x.observe(outputEl, {childList: true});\n",
              "}}\n",
              "\n",
              "                        })\n",
              "                };\n",
              "                \n",
              "            </script>\n",
              "        </div>\n",
              "</body>\n",
              "</html>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CusfZQ-SkIVn"
      },
      "source": [
        "####Split Train-Test"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "e0oU9p3BkLmz",
        "outputId": "7da6f82d-f960-4f3f-974e-8e51a89f3b47"
      },
      "source": [
        "X_train_so, X_test_so, y_train_so, y_test_so = train_test_split(x_so, label_so, test_size=0.25, random_state=1)\n",
        "print(X_train_so.shape)\n",
        "print(X_test_so.shape)\n",
        "print(y_train_so.shape)\n",
        "print(y_test_so.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(766, 3950)\n",
            "(256, 3950)\n",
            "(766,)\n",
            "(256,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "z_0cLsuBDXrN",
        "outputId": "8d653b60-ea92-4aec-c47c-cc0fee590a15"
      },
      "source": [
        "y_test_so[0]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            },
            "text/plain": [
              "'Non_R'"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0ZC8FXpl6SZ1",
        "outputId": "3a9429dd-4155-43a5-c250-e9baeb949529"
      },
      "source": [
        "lb_make = LabelEncoder()\n",
        "y_train_so = lb_make.fit_transform(y_train_so)\n",
        "print(y_train_so.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(766,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rWyFM8ha6Uj5",
        "outputId": "2c5ec1bc-2d91-4f4e-a877-7b78c03878a7"
      },
      "source": [
        "y_test_so = lb_make.fit_transform(y_test_so)\n",
        "print(y_test_so.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(256,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "denymFjJDOrT",
        "outputId": "c4fdfc07-ff3b-40c9-b2eb-508eaae0af56"
      },
      "source": [
        "y_test_so[0]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 29
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jNmAeJ27zGtb"
      },
      "source": [
        "### **Classification**\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FsoWNzJtzTOv"
      },
      "source": [
        "##### **Train**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "4pBXQQ2dDnzw"
      },
      "source": [
        "parameter_space = {\r\n",
        "    'activation': ['tanh', 'relu'],\r\n",
        "    'solver': ['sgd', 'adam'],\r\n",
        "    'alpha': [0.0001, 0.05],\r\n",
        "    'learning_rate': ['constant','adaptive'],\r\n",
        "}\r\n",
        "scoring = {'AUC': 'roc_auc', \r\n",
        "         'Accuracy': 'accuracy', \r\n",
        "         'Precision': 'precision_weighted', \r\n",
        "         'Recall':'recall_weighted', \r\n",
        "         'F1-Score':'f1_weighted'}\r\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SL3J993vmRqp",
        "outputId": "50803bbc-5ff4-4cd1-e1fb-e8dec7bb744a"
      },
      "source": [
        "k = StratifiedKFold(n_splits=10, shuffle=False)\r\n",
        "clf = GridSearchCV(MLPClassifier(max_iter=100), parameter_space, scoring=score, n_jobs=-1, cv=k,refit='AUC')\r\n",
        "start = time.time()\r\n",
        "clf.fit(X_train_so.toarray(), y_train_so)\r\n",
        "stop = time.time()\r\n",
        "print('Waktu Train ANN Over-sampling: {:.2f} s'.format(stop - start))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Waktu Train ANN Over-sampling: 1058.82 s\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5EfUPo1jSlTv",
        "outputId": "f6517ff6-8804-44e3-f5bc-0616ac53698b"
      },
      "source": [
        "print(\"Accuracy Score:\",clf.cv_results_['mean_test_Accuracy'][clf.best_index_])\r\n",
        "print(\"Precision Score:\",clf.cv_results_['mean_test_Precision'][clf.best_index_])\r\n",
        "print(\"Recall Score:\",clf.cv_results_['mean_test_Recall'][clf.best_index_])\r\n",
        "print(\"F1-Score:\",clf.cv_results_['mean_test_F1-Score'][clf.best_index_])\r\n",
        "print(\"AUC Score:\",clf.cv_results_['mean_test_AUC'][clf.best_index_])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy Score: 0.9228639781271359\n",
            "Precision Score: 0.9279678044107522\n",
            "Recall Score: 0.9228639781271359\n",
            "F1-Score: 0.9225354995912796\n",
            "AUC Score: 0.9817618438809574\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3qGrCj0mD51S",
        "outputId": "7b00852e-e48f-4544-8619-42ebd552d5c4"
      },
      "source": [
        "# Best paramete set\r\n",
        "print('Best parameters found:\\n', clf.best_estimator_)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Best parameters found:\n",
            " MLPClassifier(activation='tanh', alpha=0.0001, batch_size='auto', beta_1=0.9,\n",
            "              beta_2=0.999, early_stopping=False, epsilon=1e-08,\n",
            "              hidden_layer_sizes=(100,), learning_rate='constant',\n",
            "              learning_rate_init=0.001, max_fun=15000, max_iter=100,\n",
            "              momentum=0.9, n_iter_no_change=10, nesterovs_momentum=True,\n",
            "              power_t=0.5, random_state=None, shuffle=True, solver='adam',\n",
            "              tol=0.0001, validation_fraction=0.1, verbose=False,\n",
            "              warm_start=False)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "N1u6FimAzTOw"
      },
      "source": [
        "##### **Test**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0N__hbQczTOw"
      },
      "source": [
        "y_pred_soann = clf.predict(X_test_so.toarray())\n",
        "# y_pred_soann"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NOxTp_jlzTOx"
      },
      "source": [
        "##### **Evaluation**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wBIdMPUdzTOx",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 334
        },
        "outputId": "c5a956e1-621b-43a6-d88e-336f83606f85"
      },
      "source": [
        "#Show the Confussion Matrix\n",
        "cm1 = metrics.confusion_matrix(y_test_so, y_pred_soann)\n",
        "plot_confusion_matrix(cm1, figsize=(5,5))\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x360 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1t9XVIszzTOx",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "72ee11fb-d01b-4f46-ccd1-ecf6c908b260"
      },
      "source": [
        "print(classification_report(y_test_so, y_pred_soann))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       0.98      0.87      0.92       131\n",
            "           1       0.88      0.98      0.93       125\n",
            "\n",
            "    accuracy                           0.93       256\n",
            "   macro avg       0.93      0.93      0.93       256\n",
            "weighted avg       0.93      0.93      0.93       256\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zjI4svZIWl60",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "outputId": "9f3bd27b-c577-4c3a-d18f-185477727bad"
      },
      "source": [
        "y_pred_soann_proba = clf.predict_proba(X_test_so.toarray())[::,1]\n",
        "fprsoann, tprsoann, _ = metrics.roc_curve(y_test_so,  y_pred_soann_proba)\n",
        "aucsoann = metrics.roc_auc_score(y_test_so, y_pred_soann_proba)\n",
        "\n",
        "# ROC Curves\n",
        "plt.plot(fprsoann,tprsoann,label=\"Multi Layer Perceptron, auc=\"+str(aucsoann))\n",
        "plt.title('ROC Curve')\n",
        "plt.xlabel('false positive rate') \n",
        "plt.ylabel('true positive rate')\n",
        "plt.legend(loc=4)\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": "markdown",
      "metadata": {
        "id": "YnF_WVM8svOq"
      },
      "source": [
        "##### **User Test**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VeonhKtRsvOq",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "f2dd90e9-925d-444e-cbe8-f73ee810e251"
      },
      "source": [
        "sentence = \"\" #@param\n",
        "sentence = [sentence]\n",
        "sentence = pd.DataFrame(data=sentence,columns=['text'], index=[0])\n",
        "sentence['text'] = clean(sentence['text'])\n",
        "sentence['text'] = case_fold(sentence['text'])\n",
        "sentence['text'] = token(sentence['text'])\n",
        "sentence['text'] = stop_words(sentence['text'])\n",
        "sentence['text'] = stem(sentence['text'])\n",
        "sentence['text'] = sentence['text'].apply(lambda x: \" \".join(x) if isinstance(x, list) else x)\n",
        "clean_sentence = vec.transform(sentence['text'])\n",
        "y_pred = clf.predict(clean_sentence.toarray())\n",
        "label_pred = 'Rasis' if np.round(y_pred[0]) else 'Tidak Rasis'\n",
        "print('predicted =',label_pred)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "predicted = Tidak Rasis\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AB6D1tUFIjkN"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}