From 55d98dcca97f222751d80d504ed59a6d47a29576 Mon Sep 17 00:00:00 2001
From: Juliana Resplande <julianarsg13@gmail.com>
Date: Wed, 23 Aug 2023 10:01:21 -0300
Subject: [PATCH] Add files via upload

---
 process.ipynb | 1615 +++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 1615 insertions(+)
 create mode 100644 process.ipynb

diff --git a/process.ipynb b/process.ipynb
new file mode 100644
index 0000000..25490cc
--- /dev/null
+++ b/process.ipynb
@@ -0,0 +1,1615 @@
+{
+  "cells": [
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "!pip install matplotlib_venn gensim matpltlib pandas numpy tqdm"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 1,
+      "metadata": {
+        "id": "zIUmmW67vs3H"
+      },
+      "outputs": [],
+      "source": [
+        "from collections import Counter\n",
+        "from functools import reduce\n",
+        "import os\n",
+        "\n",
+        "\n",
+        "from matplotlib_venn import venn2_unweighted\n",
+        "from gensim.utils import deaccent\n",
+        "import matplotlib.pyplot as plt\n",
+        "from tqdm.auto import tqdm\n",
+        "import pandas as pd\n",
+        "import numpy as np\n",
+        "\n",
+        "tqdm.pandas()\n"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 2,
+      "metadata": {
+        "id": "64j4lYuov-a5"
+      },
+      "outputs": [],
+      "source": [
+        "datasets = dict()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 3,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "RAW_DATA_DIR = \"raw_data\""
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 4,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "/mnt/d/Pesquisa/Fact br eval/raw_data\n"
+          ]
+        }
+      ],
+      "source": [
+        "%cd {RAW_DATA_DIR}"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## A - Baixar dados"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "!wget https://github.com/kamplus/FakeNewsSet/raw/master/Dataset/News_fake.csv\n",
+        "!wget https://github.com/kamplus/FakeNewsSet/raw/master/Dataset/News_notFake.csv\n",
+        "!wget https://zenodo.org/record/5191798/files/central_de_fatos.csv\n",
+        "!wget https://github.com/Gabriel-Lino-Garcia/FakeRecogna/raw/master/dataset/FakeRecogna.xlsx\n",
+        "!git clone https://github.com/roneysco/Fake.br-Corpus\n",
+        "!gdown --id 1Xx_4bw37cgrEkEhQ1toLsIbAj2pLgvDx -O fact-check_tweet_dataset.zip\n",
+        "!unzip fact-check_tweet_dataset.zip"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## B - Importar dados"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 5,
+      "metadata": {
+        "id": "7ChABOSnvx5X"
+      },
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "11647\n",
+            "10551\n"
+          ]
+        }
+      ],
+      "source": [
+        "datasets[\"Central de fatos\"] = pd.read_csv(\"central_de_fatos.csv\", sep=\";\")\n",
+        "print(datasets[\"Central de fatos\"].shape[0])\n",
+        "datasets[\"Central de fatos\"][\"rating\"] = datasets[\"Central de fatos\"][\"rating\"].str.lower().apply(eval).apply(set)\n",
+        "datasets[\"Central de fatos\"] = datasets[\"Central de fatos\"][datasets[\"Central de fatos\"][\"rating\"].apply(len) == 1]\n",
+        "print(datasets[\"Central de fatos\"].shape[0])\n",
+        "datasets[\"Central de fatos\"][\"categories\"] = datasets[\"Central de fatos\"][\"categories\"] \\\n",
+        "    .apply(lambda c: eval(c) if isinstance(c, str) and \"[\" in c else c) \\\n",
+        "    .apply(lambda c: c[0] if isinstance(c, list) and len(c) > 0 else c) \\\n",
+        "    .apply(lambda c: \"\" if isinstance(c, list) and len(c) == 0 else c)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 6,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "fake_news_set_fake = pd.read_csv(\"News_fake.csv\")\n",
+        "fake_news_set_fake[\"alternative_name\"] = \"fake\"\n",
+        "\n",
+        "fake_news_set_not_fake = pd.read_csv(\"News_notFake.csv\")\n",
+        "fake_news_set_not_fake[\"alternative_name\"] = \"not_fake\"\n",
+        "\n",
+        "fake_news_set = pd.concat([fake_news_set_fake, fake_news_set_not_fake])\n",
+        "datasets[\"FakeNewsSet\"] = fake_news_set\n",
+        "datasets[\"FakeNewsSet\"][\"tweet_ids\"] = datasets[\"FakeNewsSet\"][\"tweet_ids\"].str.split(\"\\t\")"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 7,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "600\n",
+            "27059\n"
+          ]
+        }
+      ],
+      "source": [
+        "print(datasets[\"FakeNewsSet\"].shape[0])\n",
+        "print(datasets[\"FakeNewsSet\"][\"tweet_ids\"].apply(len).sum())"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 8,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "application/vnd.jupyter.widget-view+json": {
+              "model_id": "ecc093250e6c4bbab6fff9ae039dad66",
+              "version_major": 2,
+              "version_minor": 0
+            },
+            "text/plain": [
+              "  0%|          | 0/3600 [00:00<?, ?it/s]"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        },
+        {
+          "data": {
+            "application/vnd.jupyter.widget-view+json": {
+              "model_id": "bb95462184e9404998a26f00670b0f09",
+              "version_major": 2,
+              "version_minor": 0
+            },
+            "text/plain": [
+              "  0%|          | 0/3600 [00:00<?, ?it/s]"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "columns = [\n",
+        " \"author\",\n",
+        " \"link\",\n",
+        " \"category\",\n",
+        " \"date of publication\",\n",
+        " \"number of tokens\",\n",
+        " \"number of words without punctuation\",\n",
+        " \"number of types\",\n",
+        " \"number of links inside the news\",\n",
+        " \"number of words in upper case\",\n",
+        " \"number of verbs\",\n",
+        " \"number of subjuntive and imperative verbs\",\n",
+        " \"number of nouns\",\n",
+        " \"number of adjectives\",\n",
+        " \"number of adverbs\",\n",
+        " \"number of modal verbs (mainly auxiliary verbs)\",\n",
+        " \"number of singular first and second personal pronouns\",\n",
+        " \"number of plural first personal pronouns\",\n",
+        " \"number of pronouns\",\n",
+        " \"pausality\",\n",
+        " \"number of characters\",\n",
+        " \"average sentence length\",\n",
+        " \"average word length\",\n",
+        " \"percentage of news with speeling errors\",\n",
+        " \"emotiveness\",\n",
+        " \"diversity\",\n",
+        "]\n",
+        "\n",
+        "raw_data = list()\n",
+        "\n",
+        "\n",
+        "for label in [\"fake\", \"true\"]:\n",
+        "    text_dir = os.path.join(\"Fake.br-Corpus\", \"full_texts\", label)\n",
+        "    metadata_dir = os.path.join(\"Fake.br-Corpus\", \"full_texts\", f\"{label}-meta-information\")\n",
+        "    crop_dir = os.path.join(\"Fake.br-Corpus\", \"size_normalized_texts\", label)\n",
+        "\n",
+        "    for example in tqdm(os.listdir(text_dir)):\n",
+        "        text_path = os.path.join(text_dir, example)\n",
+        "        metadata_path = os.path.join(metadata_dir, example.replace(\".txt\",\"-meta.txt\"))\n",
+        "        crop_path = os.path.join(crop_dir, example)\n",
+        "    \n",
+        "        with open(text_path, \"r\") as f:\n",
+        "            text = f.read()\n",
+        "\n",
+        "        cropped_text = np.nan\n",
+        "        if os.path.exists(crop_path):\n",
+        "            with open(crop_path, \"r\") as f:\n",
+        "                cropped_text = f.read()\n",
+        "\n",
+        "        with open(metadata_path, \"r\") as f:\n",
+        "            metadata = f.read().split(\"\\n\")\n",
+        "\n",
+        "        example_data =  dict(zip(columns, metadata))\n",
+        "        example_data[\"text\"] = text\n",
+        "        example_data[\"size_normalized_text\"] = cropped_text\n",
+        "        example_data[\"label\"] = label\n",
+        "\n",
+        "        raw_data.append(example_data)\n",
+        "\n",
+        "raw_data = pd.DataFrame(raw_data)\n",
+        "prepro_data = pd.read_csv(os.path.join(\"Fake.br-Corpus\", \"preprocessed\", \"pre-processed.csv\"), index_col=\"index\")\n",
+        "raw_data[\"preprocessed_news\"] = prepro_data[\"preprocessed_news\"]\n",
+        "datasets[\"Fake.br\"] = raw_data"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 9,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "804\n",
+            "803\n"
+          ]
+        }
+      ],
+      "source": [
+        "datasets[\"Fact-check_tweet\"] = pd.read_csv(\"pt_pt.csv\")\n",
+        "print(datasets[\"Fact-check_tweet\"].shape[0])\n",
+        "datasets[\"Fact-check_tweet\"] = datasets[\"Fact-check_tweet\"][~datasets[\"Fact-check_tweet\"][\"article_url\"].str.startswith(\"https://factly.in\")]\n",
+        "datasets[\"Fact-check_tweet\"] = datasets[\"Fact-check_tweet\"].reset_index()\n",
+        "print(datasets[\"Fact-check_tweet\"].shape[0])"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 10,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "11903\n",
+            "11902\n"
+          ]
+        }
+      ],
+      "source": [
+        "datasets[\"FakeRecogna\"] = pd.read_excel(\"FakeRecogna.xlsx\")\n",
+        "print(datasets[\"FakeRecogna\"].shape[0])\n",
+        "datasets[\"FakeRecogna\"] = datasets[\"FakeRecogna\"][datasets[\"FakeRecogna\"][\"URL\"].notna()]\n",
+        "print(datasets[\"FakeRecogna\"].shape[0])"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## C - Selecionar datasets"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "Selecionei somente os datasets que possuem uma url atrelada a alegação e/ou a fonte"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 11,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "claim_datasets = {\n",
+        "    \"Fake.br\": 'link',\n",
+        "\n",
+        "    # ambos são de tweets\n",
+        "    \"Fact-check_tweet\": 'tweet_id',\n",
+        "    \"FakeNewsSet\": \"tweet_ids\",\n",
+        "}\n",
+        "\n",
+        "\n",
+        "review_datasets = {\n",
+        "    \"FakeRecogna\": 'URL',\n",
+        "    \"Central de fatos\": 'url',\n",
+        "    \"Fact-check_tweet\": 'article_url',\n",
+        "    \"FakeNewsSet\": 'news_url',\n",
+        "}"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## D - Análises iniciais"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 12,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "treat_url = lambda url: \"//\".join(url.split(\"//\")[1:]).replace(\"www.\",\"\")\n",
+        "\n",
+        "review_urls = {dataset: set(datasets[dataset][review_datasets[dataset]]) for dataset in review_datasets}\n",
+        "review_urls = {d: {treat_url(url) for url in urls} for d, urls in review_urls.items()}"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 13,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "def norm_domain(domain):\n",
+        "    for suffix in [\"gazetadigital.com.br\", \"tnh1.com.br\", \"folhavitoria.com.br\", \"jornaldebrasilia.com.br\"]:\n",
+        "        if domain in suffix:\n",
+        "            return \"r7.com\"\n",
+        "\n",
+        "    for suffix in [\"globo.com\", \"uol.com.br\", \"estadao.com.br\", \"sapo.pt\", \"afp.com\", \"r7.com\"]:\n",
+        "        if suffix in domain:\n",
+        "            return suffix\n",
+        "\n",
+        "    return domain"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 14,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "agency_count = dict()\n",
+        "\n",
+        "for dataset in review_urls.keys():\n",
+        "    dataset_agencies = Counter(url.split(\"/\")[0] for url in review_urls[dataset] if isinstance(url, str))\n",
+        "    \n",
+        "    for agency in dataset_agencies:\n",
+        "        if not agency in agency_count:\n",
+        "            agency_count[agency] = dict()\n",
+        "\n",
+        "        agency_count[agency][dataset] = dataset_agencies[agency]\n",
+        "\n",
+        "agency_count = pd.DataFrame(agency_count).T.fillna(0).astype(int)\n",
+        "\n",
+        "\n",
+        "domain_count = agency_count.reset_index(names=\"domain\")\n",
+        "domain_count[\"domain\"] = domain_count[\"domain\"].apply(norm_domain)\n",
+        "\n",
+        "domain_count = domain_count.groupby(\"domain\").sum()\n",
+        "domain_count = domain_count.loc[(domain_count > 0).sum(axis=1).sort_values(ascending=False).index]\n",
+        "domain_count = domain_count[[\"FakeRecogna\", \"Central de fatos\", \"Fact-check_tweet\", \"FakeNewsSet\"]]"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 15,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>FakeRecogna</th>\n",
+              "      <th>Central de fatos</th>\n",
+              "      <th>Fact-check_tweet</th>\n",
+              "      <th>FakeNewsSet</th>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>domain</th>\n",
+              "      <th></th>\n",
+              "      <th></th>\n",
+              "      <th></th>\n",
+              "      <th></th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>globo.com</th>\n",
+              "      <td>2366</td>\n",
+              "      <td>818</td>\n",
+              "      <td>76</td>\n",
+              "      <td>279</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>uol.com.br</th>\n",
+              "      <td>4950</td>\n",
+              "      <td>1825</td>\n",
+              "      <td>175</td>\n",
+              "      <td>138</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>afp.com</th>\n",
+              "      <td>495</td>\n",
+              "      <td>0</td>\n",
+              "      <td>49</td>\n",
+              "      <td>10</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>aosfatos.org</th>\n",
+              "      <td>0</td>\n",
+              "      <td>1431</td>\n",
+              "      <td>100</td>\n",
+              "      <td>143</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>boatos.org</th>\n",
+              "      <td>2474</td>\n",
+              "      <td>5523</td>\n",
+              "      <td>95</td>\n",
+              "      <td>0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>projetocomprova.com.br</th>\n",
+              "      <td>388</td>\n",
+              "      <td>361</td>\n",
+              "      <td>2</td>\n",
+              "      <td>0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>apublica.org</th>\n",
+              "      <td>0</td>\n",
+              "      <td>0</td>\n",
+              "      <td>4</td>\n",
+              "      <td>11</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>estadao.com.br</th>\n",
+              "      <td>0</td>\n",
+              "      <td>593</td>\n",
+              "      <td>62</td>\n",
+              "      <td>0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>e-farsas.com</th>\n",
+              "      <td>787</td>\n",
+              "      <td>0</td>\n",
+              "      <td>0</td>\n",
+              "      <td>0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>gov.br</th>\n",
+              "      <td>215</td>\n",
+              "      <td>0</td>\n",
+              "      <td>0</td>\n",
+              "      <td>0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>observador.pt</th>\n",
+              "      <td>0</td>\n",
+              "      <td>0</td>\n",
+              "      <td>21</td>\n",
+              "      <td>0</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>r7.com</th>\n",
+              "      <td>0</td>\n",
+              "      <td>0</td>\n",
+              "      <td>0</td>\n",
+              "      <td>12</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>ricmais.com.br</th>\n",
+              "      <td>0</td>\n",
+              "      <td>0</td>\n",
+              "      <td>0</td>\n",
+              "      <td>1</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>sapo.pt</th>\n",
+              "      <td>0</td>\n",
+              "      <td>0</td>\n",
+              "      <td>63</td>\n",
+              "      <td>0</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "                        FakeRecogna  Central de fatos  Fact-check_tweet  \\\n",
+              "domain                                                                    \n",
+              "globo.com                      2366               818                76   \n",
+              "uol.com.br                     4950              1825               175   \n",
+              "afp.com                         495                 0                49   \n",
+              "aosfatos.org                      0              1431               100   \n",
+              "boatos.org                     2474              5523                95   \n",
+              "projetocomprova.com.br          388               361                 2   \n",
+              "apublica.org                      0                 0                 4   \n",
+              "estadao.com.br                    0               593                62   \n",
+              "e-farsas.com                    787                 0                 0   \n",
+              "gov.br                          215                 0                 0   \n",
+              "observador.pt                     0                 0                21   \n",
+              "r7.com                            0                 0                 0   \n",
+              "ricmais.com.br                    0                 0                 0   \n",
+              "sapo.pt                           0                 0                63   \n",
+              "\n",
+              "                        FakeNewsSet  \n",
+              "domain                               \n",
+              "globo.com                       279  \n",
+              "uol.com.br                      138  \n",
+              "afp.com                          10  \n",
+              "aosfatos.org                    143  \n",
+              "boatos.org                        0  \n",
+              "projetocomprova.com.br            0  \n",
+              "apublica.org                     11  \n",
+              "estadao.com.br                    0  \n",
+              "e-farsas.com                      0  \n",
+              "gov.br                            0  \n",
+              "observador.pt                     0  \n",
+              "r7.com                           12  \n",
+              "ricmais.com.br                    1  \n",
+              "sapo.pt                           0  "
+            ]
+          },
+          "execution_count": 15,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "domain_count"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 16,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 640x480 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "ax = domain_count.T.plot(kind='bar', stacked=True, color=[plt.cm.Paired(i) for i in range(20)])\n",
+        "_ = plt.legend(bbox_to_anchor=(1.05, 1.0), loc='upper left')"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 17,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 640x480 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "_ = venn2_unweighted([review_urls[\"FakeRecogna\"], review_urls[\"Central de fatos\"]], set_labels=[\"FakeRecogna\", \"Central de fatos\"])"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 18,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "categories_1  = datasets[\"FakeRecogna\"].value_counts(subset=\"Categoria\").reset_index(name=\"FakeRecogna\")\n",
+        "categories_1[\"Categoria\"] = categories_1[\"Categoria\"].str.lower()\n",
+        "\n",
+        "categories_2 = datasets[\"Central de fatos\"][\"categories\"].value_counts().reset_index().dropna()\n",
+        "categories_2 = categories_2.rename(columns={\"index\": \"Categoria\", \"categories\": \"Central de fatos\"})\n",
+        "categories_2[\"Categoria\"] = categories_2[\"Categoria\"].str.lower()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 19,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>Categoria</th>\n",
+              "      <th>Central de fatos</th>\n",
+              "      <th>FakeRecogna</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>política</td>\n",
+              "      <td>1569</td>\n",
+              "      <td>3951</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>brasil</td>\n",
+              "      <td>1081</td>\n",
+              "      <td>904</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>saúde</td>\n",
+              "      <td>660</td>\n",
+              "      <td>4456</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>mundo</td>\n",
+              "      <td>555</td>\n",
+              "      <td>580</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>entretenimento</td>\n",
+              "      <td>464</td>\n",
+              "      <td>1409</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5</th>\n",
+              "      <td>ciência</td>\n",
+              "      <td>79</td>\n",
+              "      <td>602</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "        Categoria  Central de fatos  FakeRecogna\n",
+              "0        política              1569         3951\n",
+              "1          brasil              1081          904\n",
+              "2           saúde               660         4456\n",
+              "3           mundo               555          580\n",
+              "4  entretenimento               464         1409\n",
+              "5         ciência                79          602"
+            ]
+          },
+          "execution_count": 19,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "categories_2.merge(categories_1)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 20,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 640x480 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "_ = venn2_unweighted([review_urls[\"Fact-check_tweet\"], review_urls[\"FakeNewsSet\"]], set_labels=[\"Fact-check_tweet\", \"FakeNewsSet\"])"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 21,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "all_review_urls = reduce(lambda a, b: a.union(b), review_urls.values())\n",
+        "\n",
+        "review_urls_count = list()\n",
+        "for review_url in all_review_urls:\n",
+        "    item = {\"review_url\": review_url}\n",
+        "    item.update({dataset: review_url in review_urls[dataset] for dataset in review_urls.keys()})\n",
+        "    review_urls_count.append(item)\n",
+        "\n",
+        "review_urls_count = pd.DataFrame(review_urls_count).set_index(\"review_url\")"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 22,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "20028"
+            ]
+          },
+          "execution_count": 22,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "len(all_review_urls)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 23,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "23467"
+            ]
+          },
+          "execution_count": 23,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "domain_count.sum().sum()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 24,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>FakeRecogna</th>\n",
+              "      <th>Central de fatos</th>\n",
+              "      <th>Fact-check_tweet</th>\n",
+              "      <th>FakeNewsSet</th>\n",
+              "      <th>0</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>True</td>\n",
+              "      <td>False</td>\n",
+              "      <td>False</td>\n",
+              "      <td>False</td>\n",
+              "      <td>8841</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>False</td>\n",
+              "      <td>True</td>\n",
+              "      <td>False</td>\n",
+              "      <td>False</td>\n",
+              "      <td>7298</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>6</th>\n",
+              "      <td>False</td>\n",
+              "      <td>True</td>\n",
+              "      <td>True</td>\n",
+              "      <td>False</td>\n",
+              "      <td>192</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>False</td>\n",
+              "      <td>False</td>\n",
+              "      <td>False</td>\n",
+              "      <td>True</td>\n",
+              "      <td>312</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "   FakeRecogna  Central de fatos  Fact-check_tweet  FakeNewsSet     0\n",
+              "0         True             False             False        False  8841\n",
+              "1        False              True             False        False  7298\n",
+              "6        False              True              True        False   192\n",
+              "3        False             False             False         True   312"
+            ]
+          },
+          "execution_count": 24,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "review_urls_count.value_counts().reset_index().loc[[0,1,6,3]]"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 25,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>FakeRecogna</th>\n",
+              "      <th>Central de fatos</th>\n",
+              "      <th>Fact-check_tweet</th>\n",
+              "      <th>FakeNewsSet</th>\n",
+              "      <th>0</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>True</td>\n",
+              "      <td>True</td>\n",
+              "      <td>False</td>\n",
+              "      <td>False</td>\n",
+              "      <td>2675</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>8</th>\n",
+              "      <td>True</td>\n",
+              "      <td>False</td>\n",
+              "      <td>True</td>\n",
+              "      <td>False</td>\n",
+              "      <td>42</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>10</th>\n",
+              "      <td>True</td>\n",
+              "      <td>False</td>\n",
+              "      <td>False</td>\n",
+              "      <td>True</td>\n",
+              "      <td>5</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5</th>\n",
+              "      <td>False</td>\n",
+              "      <td>True</td>\n",
+              "      <td>False</td>\n",
+              "      <td>True</td>\n",
+              "      <td>250</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>False</td>\n",
+              "      <td>False</td>\n",
+              "      <td>True</td>\n",
+              "      <td>False</td>\n",
+              "      <td>274</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>11</th>\n",
+              "      <td>False</td>\n",
+              "      <td>False</td>\n",
+              "      <td>True</td>\n",
+              "      <td>True</td>\n",
+              "      <td>3</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "    FakeRecogna  Central de fatos  Fact-check_tweet  FakeNewsSet     0\n",
+              "2          True              True             False        False  2675\n",
+              "8          True             False              True        False    42\n",
+              "10         True             False             False         True     5\n",
+              "5         False              True             False         True   250\n",
+              "4         False             False              True        False   274\n",
+              "11        False             False              True         True     3"
+            ]
+          },
+          "execution_count": 25,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "review_urls_count.value_counts().reset_index().loc[[2,8,10,5,4,11]]"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 26,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead 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>FakeRecogna</th>\n",
+              "      <th>Central de fatos</th>\n",
+              "      <th>Fact-check_tweet</th>\n",
+              "      <th>FakeNewsSet</th>\n",
+              "      <th>0</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>7</th>\n",
+              "      <td>True</td>\n",
+              "      <td>True</td>\n",
+              "      <td>True</td>\n",
+              "      <td>False</td>\n",
+              "      <td>112</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>9</th>\n",
+              "      <td>False</td>\n",
+              "      <td>True</td>\n",
+              "      <td>True</td>\n",
+              "      <td>True</td>\n",
+              "      <td>24</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "   FakeRecogna  Central de fatos  Fact-check_tweet  FakeNewsSet    0\n",
+              "7         True              True              True        False  112\n",
+              "9        False              True              True         True   24"
+            ]
+          },
+          "execution_count": 26,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "review_urls_count.value_counts().reset_index().loc[[7,9]]"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 27,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "review_count = review_urls_count.sum(axis=1).sort_values(ascending=False).reset_index()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 28,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/plain": [
+              "1    16725\n",
+              "2     3167\n",
+              "3      136\n",
+              "Name: 0, dtype: int64"
+            ]
+          },
+          "execution_count": 28,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "review_count[0].value_counts()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 29,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "tweet_ids = dict()\n",
+        "tweet_ids[\"Fact-check_tweet\"] = set(datasets[\"Fact-check_tweet\"]['tweet_id'])\n",
+        "tweet_ids[\"FakeNewsSet\"] = reduce(lambda a, b: set(a).union(set(b)),datasets[\"FakeNewsSet\"]['tweet_ids'])"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 30,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "image/png": "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",
+            "text/plain": [
+              "<Figure size 640x480 with 1 Axes>"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "_ = venn2_unweighted([tweet_ids[\"Fact-check_tweet\"], tweet_ids[\"FakeNewsSet\"]], set_labels=[\"Fact-check_tweet\", \"FakeNewsSet\"])"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 31,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "categories_1_fixed = datasets[\"FakeRecogna\"][\"URL\"].apply(lambda url: \"//\".join(url.split(\"//\")[1:]).replace(\"www.\",\"\").split(\"/\")[1])\n",
+        "categories_1_fixed = categories_1_fixed[categories_1_fixed.str.len() < 30].dropna().value_counts().reset_index(name=\"FakeRecogna\").rename(columns={\"index\": \"Categoria\"})\n",
+        "categories_2[\"Categoria\"] = categories_2[\"Categoria\"].apply(deaccent)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 32,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "text/html": [
+              "<div>\n",
+              "<style scoped>\n",
+              "    .dataframe tbody tr th:only-of-type {\n",
+              "        vertical-align: middle;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe tbody tr th {\n",
+              "        vertical-align: top;\n",
+              "    }\n",
+              "\n",
+              "    .dataframe thead th {\n",
+              "        text-align: right;\n",
+              "    }\n",
+              "</style>\n",
+              "<table border=\"1\" class=\"dataframe\">\n",
+              "  <thead>\n",
+              "    <tr style=\"text-align: right;\">\n",
+              "      <th></th>\n",
+              "      <th>Categoria</th>\n",
+              "      <th>Central de fatos</th>\n",
+              "      <th>FakeRecogna</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td></td>\n",
+              "      <td>2418</td>\n",
+              "      <td>503</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>politica</td>\n",
+              "      <td>1569</td>\n",
+              "      <td>1549</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>brasil</td>\n",
+              "      <td>1081</td>\n",
+              "      <td>341</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>saude</td>\n",
+              "      <td>660</td>\n",
+              "      <td>1312</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>mundo</td>\n",
+              "      <td>555</td>\n",
+              "      <td>395</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5</th>\n",
+              "      <td>tecnologia</td>\n",
+              "      <td>549</td>\n",
+              "      <td>295</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>6</th>\n",
+              "      <td>entretenimento</td>\n",
+              "      <td>464</td>\n",
+              "      <td>167</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>7</th>\n",
+              "      <td>religiao</td>\n",
+              "      <td>271</td>\n",
+              "      <td>101</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>8</th>\n",
+              "      <td>esporte</td>\n",
+              "      <td>215</td>\n",
+              "      <td>276</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>9</th>\n",
+              "      <td>ciencia</td>\n",
+              "      <td>79</td>\n",
+              "      <td>39</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>10</th>\n",
+              "      <td>opiniao</td>\n",
+              "      <td>43</td>\n",
+              "      <td>2</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>11</th>\n",
+              "      <td>eleicoes</td>\n",
+              "      <td>30</td>\n",
+              "      <td>185</td>\n",
+              "    </tr>\n",
+              "  </tbody>\n",
+              "</table>\n",
+              "</div>"
+            ],
+            "text/plain": [
+              "         Categoria  Central de fatos  FakeRecogna\n",
+              "0                               2418          503\n",
+              "1         politica              1569         1549\n",
+              "2           brasil              1081          341\n",
+              "3            saude               660         1312\n",
+              "4            mundo               555          395\n",
+              "5       tecnologia               549          295\n",
+              "6   entretenimento               464          167\n",
+              "7         religiao               271          101\n",
+              "8          esporte               215          276\n",
+              "9          ciencia                79           39\n",
+              "10         opiniao                43            2\n",
+              "11        eleicoes                30          185"
+            ]
+          },
+          "execution_count": 32,
+          "metadata": {},
+          "output_type": "execute_result"
+        }
+      ],
+      "source": [
+        "categories_2.merge(categories_1_fixed,on=\"Categoria\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## E - Unificar"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 33,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "find_domain = lambda url: norm_domain(treat_url(url).split(\"/\")[0])"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 34,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "url_idx = pd.DataFrame(review_urls_count.index).reset_index()\n",
+        "url_idx[\"index\"] = url_idx[\"index\"].apply(lambda idx: f\"review_{idx:05d}\")\n",
+        "url_idx = url_idx.set_index(\"review_url\")[\"index\"].to_dict()"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 35,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "raw_df = datasets[\"FakeRecogna\"]\n",
+        "\n",
+        "fakeRecogna = pd.DataFrame()\n",
+        "\n",
+        "fakeRecogna[\"review_text\"] = raw_df[\"Titulo\"] + \"\\n\" + raw_df[\"Subtitulo\"].fillna(\"\") + \"\\n\" + raw_df[\"Noticia\"]\n",
+        "fakeRecogna[\"review_text\"] = fakeRecogna[\"review_text\"].str.strip()\n",
+        "\n",
+        "fakeRecogna[\"review_author\"] = raw_df[\"Autor\"].fillna(\"\").str.strip()\n",
+        "fakeRecogna[\"review_author\"] = fakeRecogna[\"review_author\"].apply(lambda a: a if \"/20\" not in a else \"\")\n",
+        "fakeRecogna[\"review_author\"] = fakeRecogna[\"review_author\"].fillna(\"\").str.replace(\"Por \", \"\")\n",
+        "fakeRecogna[\"review_author\"] = fakeRecogna[\"review_author\"].apply(lambda a: a if len(a.split(\" \")) <= 22 else \"\")\n",
+        "\n",
+        "fakeRecogna[\"review_url\"] = raw_df[\"URL\"].fillna(\"\").str.strip()\n",
+        "fakeRecogna[\"review_domain\"] = fakeRecogna[\"review_url\"].apply(find_domain)\n",
+        "\n",
+        "fakeRecogna[\"review_date\"] = raw_df[\"Data\"].fillna(\"\").str.strip()\n",
+        "fakeRecogna[\"review_date\"] = fakeRecogna[\"review_date\"].apply(lambda d: d if isinstance(d, str) and len(d) >= 10 else \"\")\n",
+        "fakeRecogna[\"review_date\"] = fakeRecogna[\"review_date\"].apply(lambda d: d[:10] if isinstance(d, str) and len(d) > 10 else d)\n",
+        "fakeRecogna[\"review_date\"] = fakeRecogna[\"review_date\"].fillna(\"\").str.strip()\n",
+        "\n",
+        "fakeRecogna[\"category\"] = raw_df[\"Categoria\"]\n",
+        "fakeRecogna[\"is_fake\"] = raw_df[\"Classe\"].map({0.0: 1, 1.0: -1})\n",
+        "\n",
+        "review_id = fakeRecogna[\"review_url\"].apply(lambda url: \"//\".join(url.split(\"//\")[1:]).replace(\"www.\",\"\")).map(url_idx)\n",
+        "fakeRecogna.insert(0, \"review_id\", review_id)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 36,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "FAKE_LABELS = {\"boato\", \"falso\", \"fake\", \"enganoso\", \"fora de contexto\", \"distorcido\", \"exagerado\", \"constraditório\", \"impreciso\", \"insustentável\", \"contexto errado\"}\n",
+        "\n",
+        "def map_rating(rating: set):\n",
+        "    if rating.intersection(FAKE_LABELS):\n",
+        "        return 1\n",
+        "    elif rating.intersection({\"verdadeiro, mas\", \"ainda é cedo para dizer\"}):\n",
+        "        return 0\n",
+        "    \n",
+        "    return -1\n",
+        "\n",
+        "raw_df = datasets['Central de fatos']\n",
+        "\n",
+        "\n",
+        "central_de_fatos = pd.DataFrame()\n",
+        "\n",
+        "central_de_fatos[\"review_text\"] = raw_df[\"title\"] + \"\\n\" + raw_df[\"subtitle\"].fillna(\"\") + \"\\n\" + raw_df[\"text_news\"]\n",
+        "central_de_fatos[\"review_text\"] = central_de_fatos[\"review_text\"].str.strip()\n",
+        "\n",
+        "central_de_fatos[\"review_author\"] = \"\"\n",
+        "central_de_fatos[\"review_url\"] = raw_df[\"url\"].str.strip()\n",
+        "central_de_fatos[\"review_domain\"] = central_de_fatos[\"review_url\"].apply(find_domain)\n",
+        "central_de_fatos[\"review_date\"] = raw_df[\"publication_date\"].fillna(\"\").str.strip()\n",
+        "central_de_fatos[\"category\"] = raw_df[\"categories\"].str.lower()\n",
+        "central_de_fatos[\"is_fake\"] = raw_df[\"rating\"].map(map_rating)\n",
+        "\n",
+        "review_id = central_de_fatos[\"review_url\"].apply(lambda url: \"//\".join(url.split(\"//\")[1:]).replace(\"www.\",\"\")).map(url_idx)\n",
+        "central_de_fatos.insert(0, \"review_id\", review_id)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 37,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "fact_check_tt_pt = pd.DataFrame()\n",
+        "raw_df = datasets['Fact-check_tweet']\n",
+        "fact_check_tt_pt[\"review_url\"] = raw_df[\"article_url\"]\n",
+        "fact_check_tt_pt[\"claim_ids\"] = raw_df[\"tweet_id\"].apply(lambda x: [x])\n",
+        "fact_check_tt_pt[\"is_fake\"] = raw_df[\"label\"].map({0: -1, 1:1})\n",
+        "review_id = fact_check_tt_pt[\"review_url\"].apply(lambda url: \"//\".join(url.split(\"//\")[1:]).replace(\"www.\",\"\")).map(url_idx)\n",
+        "fact_check_tt_pt.insert(0, \"review_id\", review_id)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 38,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "fakeNewsSet = pd.DataFrame()\n",
+        "raw_df = datasets[\"FakeNewsSet\"]\n",
+        "fakeNewsSet[\"review_url\"] = raw_df[\"news_url\"]\n",
+        "fakeNewsSet[\"claim_ids\"] = raw_df[\"tweet_ids\"]\n",
+        "fakeNewsSet[\"is_fake\"] = raw_df[\"alternative_name\"].map({\"fake\": 1, \"not_fake\":-1})\n",
+        "review_id = fakeNewsSet[\"review_url\"].apply(lambda url: \"//\".join(url.split(\"//\")[1:]).replace(\"www.\",\"\")).map(url_idx)\n",
+        "fakeNewsSet.insert(0, \"review_id\", review_id)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 39,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "fakebr = pd.DataFrame()\n",
+        "raw_df = datasets['Fake.br']\n",
+        "fakebr[\"claim_text\"] = raw_df[\"text\"]\n",
+        "fakebr[\"claim_author\"] = raw_df[\"author\"]\n",
+        "fakebr[\"claim_url\"] = raw_df[\"link\"]\n",
+        "fakebr[\"claim_date\"] = raw_df[\"date of publication\"]\n",
+        "fakebr[\"category\"] = raw_df[\"category\"]\n",
+        "fakebr[\"is_fake\"] = raw_df[\"label\"].map({\"fake\":  1, \"true\": -1})"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 40,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "final_dataset = {\n",
+        "    \"fact_check_tweet_pt\": fact_check_tt_pt,\n",
+        "    \"FakeNewsSet\": fakeNewsSet,\n",
+        "    \"FakeRecogna\": fakeRecogna,\n",
+        "    \"central_de_fatos\": central_de_fatos,\n",
+        "    \"fake_br\": fakebr,\n",
+        "}"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 41,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "review_urls_count.columns = [\"FakeRecogna\", \"central_de_fatos\", \"fact_check_tweet_pt\", 'FakeNewsSet']"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 42,
+      "metadata": {},
+      "outputs": [
+        {
+          "data": {
+            "application/vnd.jupyter.widget-view+json": {
+              "model_id": "4b18af638921437b803f6099346ada8b",
+              "version_major": 2,
+              "version_minor": 0
+            },
+            "text/plain": [
+              "  0%|          | 0/3303 [00:00<?, ?it/s]"
+            ]
+          },
+          "metadata": {},
+          "output_type": "display_data"
+        }
+      ],
+      "source": [
+        "diff_reviews = set()\n",
+        "\n",
+        "intersection_urls = review_urls_count[review_urls_count.sum(axis=1) > 1]\n",
+        "for url, row in tqdm(intersection_urls.iterrows(), total=intersection_urls.shape[0]):\n",
+        "    intersection_label = set()\n",
+        "\n",
+        "    review_id = url_idx[url]\n",
+        "    for dataset in row[row == True].index:\n",
+        "        dataset_label = final_dataset[dataset][final_dataset[dataset][\"review_id\"] == review_id][\"is_fake\"].iloc[0]\n",
+        "        intersection_label.add(dataset_label)\n",
+        "\n",
+        "    if len(intersection_label) > 1:\n",
+        "        diff_reviews.add(review_id)\n",
+        "    "
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 43,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "3303 240\n"
+          ]
+        }
+      ],
+      "source": [
+        "print(intersection_urls.shape[0], len(diff_reviews))"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 44,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "for dataset in final_dataset:\n",
+        "    if not \"review_id\" in final_dataset[dataset].columns:\n",
+        "        continue\n",
+        "\n",
+        "    final_dataset[dataset] = final_dataset[dataset][~final_dataset[dataset][\"review_id\"].isin(diff_reviews)]"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 45,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "/mnt/d/Pesquisa/Fact br eval\n"
+          ]
+        }
+      ],
+      "source": [
+        "%cd .."
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 46,
+      "metadata": {},
+      "outputs": [
+        {
+          "ename": "FileExistsError",
+          "evalue": "[Errno 17] File exists: 'data'",
+          "output_type": "error",
+          "traceback": [
+            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+            "\u001b[0;31mFileExistsError\u001b[0m                           Traceback (most recent call last)",
+            "Cell \u001b[0;32mIn[46], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m os\u001b[39m.\u001b[39;49mmakedirs(\u001b[39m\"\u001b[39;49m\u001b[39mdata\u001b[39;49m\u001b[39m\"\u001b[39;49m)\n",
+            "File \u001b[0;32m<frozen os>:225\u001b[0m, in \u001b[0;36mmakedirs\u001b[0;34m(name, mode, exist_ok)\u001b[0m\n",
+            "\u001b[0;31mFileExistsError\u001b[0m: [Errno 17] File exists: 'data'"
+          ]
+        }
+      ],
+      "source": [
+        "os.makedirs(\"data\")"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 49,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "for dataset in final_dataset:\n",
+        "    final_dataset[dataset].to_csv(os.path.join(\"data\",f\"{dataset}.tsv\"), sep=\"\\t\", index=None)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": 52,
+      "metadata": {},
+      "outputs": [
+        {
+          "name": "stdout",
+          "output_type": "stream",
+          "text": [
+            "updating: data/ (stored 0%)\n",
+            "updating: data/central_de_fatos.tsv (deflated 68%)\n",
+            "updating: data/fact_check_tweet_pt.tsv (deflated 72%)\n",
+            "updating: data/FakeNewsSet.tsv (deflated 61%)\n",
+            "updating: data/FakeRecogna.tsv (deflated 65%)\n",
+            "updating: data/fake_br.tsv (deflated 63%)\n"
+          ]
+        }
+      ],
+      "source": [
+        "!zip FactChecksbr.zip -r data"
+      ]
+    }
+  ],
+  "metadata": {
+    "colab": {
+      "provenance": []
+    },
+    "kernelspec": {
+      "display_name": "Python 3",
+      "name": "python3"
+    },
+    "language_info": {
+      "codemirror_mode": {
+        "name": "ipython",
+        "version": 3
+      },
+      "file_extension": ".py",
+      "mimetype": "text/x-python",
+      "name": "python",
+      "nbconvert_exporter": "python",
+      "pygments_lexer": "ipython3",
+      "version": "3.11.4"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 0
+}