Fake_News_Detection_with_LSTM_in_Tensorflow.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4",
      "authorship_tag": "ABX9TyOmO4YqjUAaArHMl3UzDQEP",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/bchaoss/Fake-News-Detection/blob/main/Fake_News_Detection_with_LSTM_in_Tensorflow.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Fake News Detection with LSTM Model in Tensorflow2.x"
      ],
      "metadata": {
        "id": "0qSQQTx4JLHx"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "This notebook mainly focuses on building an LSTM model for fake news detection. It is modified from this great work (https://github.com/matdekoning/FakeNewsClassifier). The model utilizes a dataset available on Kaggle, which can be downloaded from this link: https://www.kaggle.com/c/fake-news/data.\n",
        "\n",
        "In this work, we leverages the LSTM model to classify news articles and determine their authenticity. LSTM (Long Short-Term Memory) is a type of recurrent neural network that can capture contextual information in text, aiding in better understanding the content and semantics of news articles. By training on a large dataset of real and fake news, the model learns key features to classify new articles."
      ],
      "metadata": {
        "id": "SnbmYfaFHc8_"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Install data from kaggle"
      ],
      "metadata": {
        "id": "n5UHnYlsJZA_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# connext kaggle before run this cell\n",
        "!kaggle competitions download -c fake-news\n",
        "!kaggle datasets download -d danielwillgeorge/glove6b100dtxt\n",
        "\n",
        "!unzip /content/competitions/fake-news/fake-news.zip\n",
        "!unzip /content/datasets/danielwillgeorge/glove6b100dtxt/glove6b100dtxt.zip"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "rLCvC-BlCEcJ",
        "outputId": "c7a1bbf2-c81a-47e2-997f-2856e243b050"
      },
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading fake-news.zip to /content/competitions/fake-news\n",
            " 88% 41.0M/46.5M [00:02<00:00, 28.3MB/s]\n",
            "100% 46.5M/46.5M [00:02<00:00, 21.0MB/s]\n",
            "Downloading glove6b100dtxt.zip to /content/datasets/danielwillgeorge/glove6b100dtxt\n",
            " 94% 124M/131M [00:01<00:00, 122MB/s]\n",
            "100% 131M/131M [00:01<00:00, 114MB/s]\n",
            "Archive:  /content/competitions/fake-news/fake-news.zip\n",
            "  inflating: submit.csv              \n",
            "  inflating: test.csv                \n",
            "  inflating: train.csv               \n",
            "Archive:  /content/datasets/danielwillgeorge/glove6b100dtxt/glove6b100dtxt.zip\n",
            "  inflating: glove.6B.100d.txt       \n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import matplotlib.ticker as ticker\n",
        "import os\n",
        "\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras import layers\n",
        "from tensorflow.keras.models import Sequential\n",
        "\n",
        "from tensorflow.keras.preprocessing.text import Tokenizer\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import confusion_matrix, f1_score"
      ],
      "metadata": {
        "id": "nJ6mPPzqBi5O"
      },
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Prepare the Data"
      ],
      "metadata": {
        "id": "imTTY4TgJun3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# choose a availble GPU device\n",
        "physical_devices = tf.config.list_physical_devices('GPU')\n",
        "tf.config.experimental.set_memory_growth(physical_devices[0], True)"
      ],
      "metadata": {
        "id": "X3wrAbytBn37"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# read the train/test data\n",
        "train_df = pd.read_csv('/content/train.csv')\n",
        "test_df = pd.read_csv('/content/test.csv')\n",
        "datasets = [train_df, test_df]"
      ],
      "metadata": {
        "id": "QbzsfCIQBqB-"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "Gcm0mOi7C0NI",
        "outputId": "cdb488f1-5697-49c7-f919-6801c8e66c21"
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   id                                              title              author  \\\n",
              "0   0  House Dem Aide: We Didn’t Even See Comey’s Let...       Darrell Lucus   \n",
              "1   1  FLYNN: Hillary Clinton, Big Woman on Campus - ...     Daniel J. Flynn   \n",
              "2   2                  Why the Truth Might Get You Fired  Consortiumnews.com   \n",
              "3   3  15 Civilians Killed In Single US Airstrike Hav...     Jessica Purkiss   \n",
              "4   4  Iranian woman jailed for fictional unpublished...      Howard Portnoy   \n",
              "\n",
              "                                                text  label  \n",
              "0  House Dem Aide: We Didn’t Even See Comey’s Let...      1  \n",
              "1  Ever get the feeling your life circles the rou...      0  \n",
              "2  Why the Truth Might Get You Fired October 29, ...      1  \n",
              "3  Videos 15 Civilians Killed In Single US Airstr...      1  \n",
              "4  Print \\nAn Iranian woman has been sentenced to...      1  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-4c7b92d9-9c3f-4b11-9bec-15e50bdfe141\" class=\"colab-df-container\">\n",
              "    <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>id</th>\n",
              "      <th>title</th>\n",
              "      <th>author</th>\n",
              "      <th>text</th>\n",
              "      <th>label</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>House Dem Aide: We Didn’t Even See Comey’s Let...</td>\n",
              "      <td>Darrell Lucus</td>\n",
              "      <td>House Dem Aide: We Didn’t Even See Comey’s Let...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>FLYNN: Hillary Clinton, Big Woman on Campus - ...</td>\n",
              "      <td>Daniel J. Flynn</td>\n",
              "      <td>Ever get the feeling your life circles the rou...</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>Why the Truth Might Get You Fired</td>\n",
              "      <td>Consortiumnews.com</td>\n",
              "      <td>Why the Truth Might Get You Fired October 29, ...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>15 Civilians Killed In Single US Airstrike Hav...</td>\n",
              "      <td>Jessica Purkiss</td>\n",
              "      <td>Videos 15 Civilians Killed In Single US Airstr...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>Iranian woman jailed for fictional unpublished...</td>\n",
              "      <td>Howard Portnoy</td>\n",
              "      <td>Print \\nAn Iranian woman has been sentenced to...</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4c7b92d9-9c3f-4b11-9bec-15e50bdfe141')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-4c7b92d9-9c3f-4b11-9bec-15e50bdfe141 button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-4c7b92d9-9c3f-4b11-9bec-15e50bdfe141');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-4be7bed2-e9c8-4057-a159-85f3e07b8610\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-4be7bed2-e9c8-4057-a159-85f3e07b8610')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-4be7bed2-e9c8-4057-a159-85f3e07b8610 button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_df.isnull().sum()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "43XOxj18C016",
        "outputId": "72ddbb87-a284-4bfd-ce46-4065b09511e6"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "id           0\n",
              "title      558\n",
              "author    1957\n",
              "text        39\n",
              "label        0\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_df.dropna(subset=['text'], inplace=True)"
      ],
      "metadata": {
        "id": "xReEdUeQC3sE"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Since NaN + string = NaN, we must fill NaN with empty string first.\n",
        "def get_full_text(df):\n",
        "    df.fillna('', inplace=True)     # fill NaN\n",
        "    df['clean_news'] = df['title'] + ' ' + df['text']\n",
        "    df['clean_news'] = df['clean_news'].str.lower().str.replace('[^A-Za-z0-9\\s]', '')\n",
        "    # no return, change df directly\n",
        "\n",
        "for df in datasets:\n",
        "    get_full_text(df)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PSf8OxsgC7Ul",
        "outputId": "b1da8305-898a-459b-c271-75c8fdf8eca7"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-12-ce2fd0bd3ef9>:5: FutureWarning: The default value of regex will change from True to False in a future version.\n",
            "  df['clean_news'] = df['clean_news'].str.lower().str.replace('[^A-Za-z0-9\\s]', '')\n",
            "<ipython-input-12-ce2fd0bd3ef9>:5: FutureWarning: The default value of regex will change from True to False in a future version.\n",
            "  df['clean_news'] = df['clean_news'].str.lower().str.replace('[^A-Za-z0-9\\s]', '')\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_df.isnull().sum()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "RaOTtHb5C_wa",
        "outputId": "21fa8062-2fcf-4535-ac0a-704ac13d84d0"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "id            0\n",
              "title         0\n",
              "author        0\n",
              "text          0\n",
              "label         0\n",
              "clean_news    0\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "count_empty_strings = len(train_df[train_df['clean_news'] == ' '])\n",
        "print(\"number of empty strings：\", count_empty_strings)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "28yJWjjMDB0S",
        "outputId": "6ba875bc-bee5-46d9-b977-385e2775b3cd"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "number of empty strings： 1\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import nltk\n",
        "nltk.download('stopwords')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UYLFpVsjILi6",
        "outputId": "c220bf0b-a0d2-4e98-9780-54649d6a7dda"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Unzipping corpora/stopwords.zip.\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def remove_stopwords(df, stop_words):\n",
        "    df['clean_news'] = df['clean_news'].apply(lambda x:\" \".join([word for word in x.split() if word not in stop_words]))\n",
        "\n",
        "from nltk.corpus import stopwords\n",
        "stop = stopwords.words('english')\n",
        "for df in datasets:\n",
        "    remove_stopwords(df, stop)"
      ],
      "metadata": {
        "id": "9RtEVyf3DFPn"
      },
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Preprocess train/test data in the same time\n",
        "def preprocess_data(train_df, test_df, text_column, max_length):\n",
        "    # merge the text data of the train and test sets\n",
        "#     all_texts = train_df[text_column].tolist() + test_df[text_column].tolist()\n",
        "    all_texts = train_df[text_column].tolist()\n",
        "\n",
        "    # build the Tokenizer\n",
        "    tokenizer = Tokenizer(oov_token='<OOV>')\n",
        "    tokenizer.fit_on_texts(all_texts)\n",
        "\n",
        "    # convert the text data into integer sequences\n",
        "    train_sequences = tokenizer.texts_to_sequences(train_df[text_column])\n",
        "    test_sequences = tokenizer.texts_to_sequences(test_df[text_column])\n",
        "\n",
        "    # perform padding on the sequences, padding to the specified length\n",
        "    train_padded_sequences = pad_sequences(train_sequences, maxlen=max_length, padding='post',truncating='post')\n",
        "    test_padded_sequences = pad_sequences(test_sequences, maxlen=max_length, padding='post',truncating='post')\n",
        "\n",
        "    return train_padded_sequences, test_padded_sequences, tokenizer"
      ],
      "metadata": {
        "id": "ZkxS_rY_DK6q"
      },
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "max_length = 600\n",
        "train_padded_sequences, test_padded_sequences, tokenizer = preprocess_data(train_df, test_df, 'clean_news', max_length)\n",
        "\n",
        "# get word index and vocabulary size\n",
        "word_index = tokenizer.word_index\n",
        "vocab_size = len(word_index) + 1"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "D4DMwPrKDSQ7",
        "outputId": "7752dfc2-67ff-403b-b38c-9cd471756d63"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "CPU times: user 11.5 s, sys: 80.4 ms, total: 11.6 s\n",
            "Wall time: 11.6 s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# get embedding dictionary from glove with dim=100\n",
        "def get_embedding_dict():\n",
        "    embedding_dict = {}\n",
        "    with open('/content/glove.6B.100d.txt', encoding='utf-8') as f:\n",
        "        for line in f:\n",
        "            values = line.split()\n",
        "            word = values[0]\n",
        "            coefs = np.asarray(values[1:], dtype='float32')\n",
        "            embedding_dict[word] = coefs\n",
        "\n",
        "    return embedding_dict\n",
        "\n",
        "embedding_dict = get_embedding_dict()"
      ],
      "metadata": {
        "id": "y47VEVFnD3XX"
      },
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# get embedding martrix for initializing the embedding layer\n",
        "embedding_dim = 100\n",
        "\n",
        "def get_embedding_matrix(vocab_size, embedding_dim, word_index, embedding_dict):\n",
        "    embedding_matrix = np.zeros((vocab_size, embedding_dim))\n",
        "    for word, i in word_index.items():\n",
        "        embedding_vector = embedding_dict.get(word)\n",
        "        if embedding_vector is not None:\n",
        "            embedding_matrix[i] = embedding_vector\n",
        "\n",
        "    return embedding_matrix\n",
        "\n",
        "embedding_matrix = get_embedding_matrix(vocab_size, embedding_dim, word_index, embedding_dict)"
      ],
      "metadata": {
        "id": "PLJ_uL2qENWk"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# split the train dataset\n",
        "X_train, X_val, Y_train, Y_val = train_test_split(train_padded_sequences, train_df['label'], test_size=0.2, random_state=42, stratify=train_df['label'])\n",
        "X_train.shape, Y_train.shape, X_val.shape, Y_val.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Cnf2DODoEyiC",
        "outputId": "aaa34f0f-9141-4798-9695-ae7224b4b583"
      },
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "((16608, 600), (16608,), (4153, 600), (4153,))"
            ]
          },
          "metadata": {},
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### MODEL"
      ],
      "metadata": {
        "id": "zuq98e3fJ2_6"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model_lstm = Sequential([\n",
        "                layers.Embedding(vocab_size, embedding_dim, weights=[embedding_matrix], trainable=False),\n",
        "                layers.Dropout(0.2),\n",
        "                layers.Conv1D(64, 5, activation='relu'),\n",
        "                layers.MaxPooling1D(pool_size=4),\n",
        "                layers.LSTM(32, return_sequences=True),\n",
        "                layers.LSTM(32),\n",
        "                # layers.GlobalAveragePooling1D(),\n",
        "                layers.Dense(128),\n",
        "                layers.Dense(1, activation='sigmoid')\n",
        "])\n",
        "\n",
        "model_lstm.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6iuwR4owE1F5",
        "outputId": "35b7fd55-493e-4228-b0e8-4bb378862abe"
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " embedding_1 (Embedding)     (None, None, 100)         18842700  \n",
            "                                                                 \n",
            " dropout_1 (Dropout)         (None, None, 100)         0         \n",
            "                                                                 \n",
            " conv1d_1 (Conv1D)           (None, None, 64)          32064     \n",
            "                                                                 \n",
            " max_pooling1d_1 (MaxPoolin  (None, None, 64)          0         \n",
            " g1D)                                                            \n",
            "                                                                 \n",
            " lstm_2 (LSTM)               (None, None, 32)          12416     \n",
            "                                                                 \n",
            " lstm_3 (LSTM)               (None, 32)                8320      \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 128)               4224      \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 1)                 129       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 18899853 (72.10 MB)\n",
            "Trainable params: 57153 (223.25 KB)\n",
            "Non-trainable params: 18842700 (71.88 MB)\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model_lstm.compile(loss='binary_crossentropy', optimizer='adam', metrics='accuracy')"
      ],
      "metadata": {
        "id": "9RrYlf-3E3TN"
      },
      "execution_count": 25,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# convert data to tensor for trainning\n",
        "X_train_tf = tf.convert_to_tensor(X_train)\n",
        "Y_train_tf = tf.convert_to_tensor(Y_train)\n",
        "X_val_tf = tf.convert_to_tensor(X_val)\n",
        "Y_val_tf = tf.convert_to_tensor(Y_val)\n",
        "\n",
        "# trainning on GPU\n",
        "with tf.device('/GPU:0'):\n",
        "    history1 = model_lstm.fit(X_train_tf, Y_train_tf, epochs=8, batch_size=100, validation_data=(X_val_tf, Y_val_tf))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hXP2IbSEE5DN",
        "outputId": "845f1467-9e3f-4cb6-8042-7f4dcbb91839"
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/8\n",
            "167/167 [==============================] - 19s 25ms/step - loss: 0.5281 - accuracy: 0.7410 - val_loss: 0.3763 - val_accuracy: 0.8375\n",
            "Epoch 2/8\n",
            "167/167 [==============================] - 3s 17ms/step - loss: 0.3031 - accuracy: 0.8754 - val_loss: 0.2383 - val_accuracy: 0.9085\n",
            "Epoch 3/8\n",
            "167/167 [==============================] - 3s 17ms/step - loss: 0.2042 - accuracy: 0.9225 - val_loss: 0.1912 - val_accuracy: 0.9294\n",
            "Epoch 4/8\n",
            "167/167 [==============================] - 3s 17ms/step - loss: 0.1256 - accuracy: 0.9555 - val_loss: 0.1121 - val_accuracy: 0.9624\n",
            "Epoch 5/8\n",
            "167/167 [==============================] - 3s 20ms/step - loss: 0.0836 - accuracy: 0.9721 - val_loss: 0.0884 - val_accuracy: 0.9656\n",
            "Epoch 6/8\n",
            "167/167 [==============================] - 3s 17ms/step - loss: 0.0639 - accuracy: 0.9790 - val_loss: 0.1159 - val_accuracy: 0.9653\n",
            "Epoch 7/8\n",
            "167/167 [==============================] - 3s 17ms/step - loss: 0.0565 - accuracy: 0.9819 - val_loss: 0.0969 - val_accuracy: 0.9757\n",
            "Epoch 8/8\n",
            "167/167 [==============================] - 3s 17ms/step - loss: 0.0523 - accuracy: 0.9836 - val_loss: 0.0882 - val_accuracy: 0.9723\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Trainning Result:"
      ],
      "metadata": {
        "id": "eMksfjX1J6zQ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def plot_trainning_result(mode='accuracy'):\n",
        "    epochs_num = len(history1.history[f'{mode}'])\n",
        "    plt.plot(history1.history[f'{mode}'])\n",
        "    plt.plot(history1.history[f'val_{mode}'])\n",
        "    plt.xlabel('epochs')\n",
        "    plt.ylabel(f'{mode}')\n",
        "    plt.legend(['train', 'val'])\n",
        "    plt.title(f'training {mode}')\n",
        "\n",
        "    ax = plt.gca()\n",
        "    ax.xaxis.set_major_locator(ticker.MaxNLocator(integer=True))\n",
        "    ax.yaxis.set_major_locator(ticker.MaxNLocator(integer=True))\n",
        "    plt.xticks(list(range(epochs_num)), list(range(1, epochs_num+1)))\n",
        "\n",
        "    plt.show()\n",
        "\n",
        "plot_trainning_result('accuracy')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "_fXz_sZsFhxL",
        "outputId": "cca53bb0-8901-4b80-b85c-d12958fb4b21"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "plot_trainning_result('loss')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "OHSTkZ3wGfFA",
        "outputId": "74f576d2-870b-4450-8800-edce190cfa3e"
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Predicting"
      ],
      "metadata": {
        "id": "s__EF8qaJ-3C"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "test_pred = model_lstm.predict(test_padded_sequences, batch_size=64, verbose=1)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4fZDKGteGkrb",
        "outputId": "faf8f9b6-cd77-44ce-c83b-77cbb1def27f"
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "82/82 [==============================] - 1s 8ms/step\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "test_df = pd.DataFrame({'id': test_df['id'].to_list(), 'label': test_pred.reshape((5200,))})\n",
        "\n",
        "test_df['label'].hist()\n",
        "plt.title('Distribution of Predicted Labels') # due to sigmoid function\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 452
        },
        "id": "84WxoXASGqZi",
        "outputId": "4f47d2aa-1928-4a84-912a-47014e6a4733"
      },
      "execution_count": 30,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# classifier\n",
        "threshold = 0.5\n",
        "test_df['label'] = np.where(test_df['label'] >= threshold, 1, 0)\n",
        "test_df['label'].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iSKva8WBGsSQ",
        "outputId": "a557156f-d059-45c5-f38c-d04059ba8c7a"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    2665\n",
              "1    2535\n",
              "Name: label, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Evaluation"
      ],
      "metadata": {
        "id": "Acll_9Gv113y"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "predictions = model_lstm.predict(train_padded_sequences, batch_size=64, verbose=1)\n",
        "\n",
        "pred_df = pd.DataFrame({'label': predictions.reshape((len(predictions),))})\n",
        "\n",
        "threshold = 0.5\n",
        "pred_df['label'] = np.where(pred_df['label'] >= threshold, 1, 0)\n",
        "pred_df['label'].value_counts()"
      ],
      "metadata": {
        "id": "Ipi2KDSo2bHo",
        "outputId": "1fd1f937-7966-404f-d6e8-24732ab1643c",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 60,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "325/325 [==============================] - 2s 7ms/step\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0    10485\n",
              "1    10276\n",
              "Name: label, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 60
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "true_labels = train_df['label']\n",
        "\n",
        "cm = confusion_matrix(true_labels, pred_df['label'])\n",
        "print(\"Confusion Matrix:\")\n",
        "print(cm)"
      ],
      "metadata": {
        "id": "JlNyd7ct13Fw",
        "outputId": "7334289a-d462-42dc-8735-fd4f175ea82b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 61,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Confusion Matrix:\n",
            "[[10312    75]\n",
            " [  173 10201]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "f1 = f1_score(true_labels, pred_df['label'])\n",
        "print(\"F1 Score:\", f1)"
      ],
      "metadata": {
        "id": "AiwtNl3C2G1k",
        "outputId": "5df6cd9a-ed61-4fd5-ac88-b01514ace285",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "F1 Score: 0.9879903147699757\n"
          ]
        }
      ]
    }
  ]
}