8 Practical.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Practical"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Setting the Scene\n",
    "\n",
    "Imagine you are working for a social media company. The company is concerned with the growing amount of fake news circulatng on it's platform. They have assigned you to investigate how fake news can be recognized and create a method of identifying it. Let's work through this problem together, first by exploring and cleaning the data and then working to classify fake vs factual news stories. We'll also create some plots of our outputs and discuss how we would communicate our findings to stakeholders."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# set plot options\n",
    "plt.rcParams['figure.figsize'] = (12, 8)\n",
    "default_plot_colour = \"#00bfbf\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_csv(\"fake_news_data.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>title</th>\n",
       "      <th>text</th>\n",
       "      <th>date</th>\n",
       "      <th>fake_or_factual</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>HOLLYWEIRD LIB SUSAN SARANDON Compares Muslim ...</td>\n",
       "      <td>There are two small problems with your analogy...</td>\n",
       "      <td>Dec 30, 2015</td>\n",
       "      <td>Fake News</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Elijah Cummings Called Trump Out To His Face ...</td>\n",
       "      <td>Buried in Trump s bonkers interview with New Y...</td>\n",
       "      <td>April 6, 2017</td>\n",
       "      <td>Fake News</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Hillary Clinton Says Half Her Cabinet Will Be...</td>\n",
       "      <td>Women make up over 50 percent of this country,...</td>\n",
       "      <td>April 26, 2016</td>\n",
       "      <td>Fake News</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Russian bombing of U.S.-backed forces being di...</td>\n",
       "      <td>WASHINGTON (Reuters) - U.S. Defense Secretary ...</td>\n",
       "      <td>September 18, 2017</td>\n",
       "      <td>Factual News</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Britain says window to restore Northern Irelan...</td>\n",
       "      <td>BELFAST (Reuters) - Northern Ireland s politic...</td>\n",
       "      <td>September 4, 2017</td>\n",
       "      <td>Factual News</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               title  \\\n",
       "0  HOLLYWEIRD LIB SUSAN SARANDON Compares Muslim ...   \n",
       "1   Elijah Cummings Called Trump Out To His Face ...   \n",
       "2   Hillary Clinton Says Half Her Cabinet Will Be...   \n",
       "3  Russian bombing of U.S.-backed forces being di...   \n",
       "4  Britain says window to restore Northern Irelan...   \n",
       "\n",
       "                                                text                 date  \\\n",
       "0  There are two small problems with your analogy...         Dec 30, 2015   \n",
       "1  Buried in Trump s bonkers interview with New Y...        April 6, 2017   \n",
       "2  Women make up over 50 percent of this country,...       April 26, 2016   \n",
       "3  WASHINGTON (Reuters) - U.S. Defense Secretary ...  September 18, 2017    \n",
       "4  BELFAST (Reuters) - Northern Ireland s politic...   September 4, 2017    \n",
       "\n",
       "  fake_or_factual  \n",
       "0       Fake News  \n",
       "1       Fake News  \n",
       "2       Fake News  \n",
       "3    Factual News  \n",
       "4    Factual News  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 198 entries, 0 to 197\n",
      "Data columns (total 4 columns):\n",
      " #   Column           Non-Null Count  Dtype \n",
      "---  ------           --------------  ----- \n",
      " 0   title            198 non-null    object\n",
      " 1   text             198 non-null    object\n",
      " 2   date             198 non-null    object\n",
      " 3   fake_or_factual  198 non-null    object\n",
      "dtypes: object(4)\n",
      "memory usage: 6.3+ KB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Classification')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot number of fake and factual articles\n",
    "data['fake_or_factual'].value_counts().plot(kind='bar', color=default_plot_colour)\n",
    "plt.title('Count of Article Classification')\n",
    "plt.ylabel('# of Articles')\n",
    "plt.xlabel('Classification')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import packages required for processing and analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "import spacy\n",
    "from spacy import displacy\n",
    "from spacy import tokenizer\n",
    "import re\n",
    "import nltk\n",
    "from nltk.tokenize import word_tokenize\n",
    "from nltk.stem import PorterStemmer, WordNetLemmatizer\n",
    "from nltk.corpus import stopwords\n",
    "from vaderSentiment.vaderSentiment import SentimentIntensityAnalyzer\n",
    "import gensim\n",
    "import gensim.corpora as corpora\n",
    "from gensim.models.coherencemodel import CoherenceModel\n",
    "from gensim.models import LsiModel, TfidfModel\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression, SGDClassifier\n",
    "from sklearn.metrics import accuracy_score, classification_report"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## POS Tagging"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "nlp = spacy.load('en_core_web_sm')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# split data by fake and factual news\n",
    "fake_news = data[data['fake_or_factual'] == \"Fake News\"]\n",
    "fact_news = data[data['fake_or_factual'] == \"Factual News\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create spacey documents - use pipe for dataframe\n",
    "fake_spaceydocs = list(nlp.pipe(fake_news['text']))\n",
    "fact_spaceydocs = list(nlp.pipe(fact_news['text'])) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create function to extract tags for each document in our data\n",
    "def extract_token_tags(doc:spacy.tokens.doc.Doc):\n",
    "    return [(i.text, i.ent_type_, i.pos_) for i in doc]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# tag fake dataset \n",
    "fake_tagsdf = []\n",
    "columns = [\"token\", \"ner_tag\", \"pos_tag\"]\n",
    "\n",
    "for ix, doc in enumerate(fake_spaceydocs):\n",
    "    tags = extract_token_tags(doc)\n",
    "    tags = pd.DataFrame(tags)\n",
    "    tags.columns = columns\n",
    "    fake_tagsdf.append(tags)\n",
    "        \n",
    "fake_tagsdf = pd.concat(fake_tagsdf)   \n",
    "\n",
    "# tag factual dataset \n",
    "fact_tagsdf = []\n",
    "\n",
    "for ix, doc in enumerate(fact_spaceydocs):\n",
    "    tags = extract_token_tags(doc)\n",
    "    tags = pd.DataFrame(tags)\n",
    "    tags.columns = columns\n",
    "    fact_tagsdf.append(tags)\n",
    "        \n",
    "fact_tagsdf = pd.concat(fact_tagsdf)   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>token</th>\n",
       "      <th>ner_tag</th>\n",
       "      <th>pos_tag</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>There</td>\n",
       "      <td></td>\n",
       "      <td>PRON</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>are</td>\n",
       "      <td></td>\n",
       "      <td>VERB</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>two</td>\n",
       "      <td>CARDINAL</td>\n",
       "      <td>NUM</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>small</td>\n",
       "      <td></td>\n",
       "      <td>ADJ</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>problems</td>\n",
       "      <td></td>\n",
       "      <td>NOUN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      token   ner_tag pos_tag\n",
       "0     There              PRON\n",
       "1       are              VERB\n",
       "2       two  CARDINAL     NUM\n",
       "3     small               ADJ\n",
       "4  problems              NOUN"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fake_tagsdf.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>token</th>\n",
       "      <th>pos_tag</th>\n",
       "      <th>counts</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>,</td>\n",
       "      <td>PUNCT</td>\n",
       "      <td>1908</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7451</th>\n",
       "      <td>the</td>\n",
       "      <td>DET</td>\n",
       "      <td>1834</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>.</td>\n",
       "      <td>PUNCT</td>\n",
       "      <td>1530</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5766</th>\n",
       "      <td>of</td>\n",
       "      <td>ADP</td>\n",
       "      <td>922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2665</th>\n",
       "      <td>and</td>\n",
       "      <td>CCONJ</td>\n",
       "      <td>875</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2449</th>\n",
       "      <td>a</td>\n",
       "      <td>DET</td>\n",
       "      <td>805</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td></td>\n",
       "      <td>SPACE</td>\n",
       "      <td>795</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7528</th>\n",
       "      <td>to</td>\n",
       "      <td>PART</td>\n",
       "      <td>767</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4921</th>\n",
       "      <td>in</td>\n",
       "      <td>ADP</td>\n",
       "      <td>668</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5099</th>\n",
       "      <td>is</td>\n",
       "      <td>AUX</td>\n",
       "      <td>419</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     token pos_tag  counts\n",
       "29       ,   PUNCT    1908\n",
       "7451   the     DET    1834\n",
       "41       .   PUNCT    1530\n",
       "5766    of     ADP     922\n",
       "2665   and   CCONJ     875\n",
       "2449     a     DET     805\n",
       "0            SPACE     795\n",
       "7528    to    PART     767\n",
       "4921    in     ADP     668\n",
       "5099    is     AUX     419"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# token frequency count (fake)\n",
    "pos_counts_fake = fake_tagsdf.groupby(['token','pos_tag']).size().reset_index(name='counts').sort_values(by='counts', ascending=False)\n",
    "pos_counts_fake.head(10)"
   ]
  },
  {
   "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>token</th>\n",
       "      <th>pos_tag</th>\n",
       "      <th>counts</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>6145</th>\n",
       "      <td>the</td>\n",
       "      <td>DET</td>\n",
       "      <td>1903</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>,</td>\n",
       "      <td>PUNCT</td>\n",
       "      <td>1698</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>.</td>\n",
       "      <td>PUNCT</td>\n",
       "      <td>1382</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4716</th>\n",
       "      <td>of</td>\n",
       "      <td>ADP</td>\n",
       "      <td>884</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1898</th>\n",
       "      <td>a</td>\n",
       "      <td>DET</td>\n",
       "      <td>789</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2093</th>\n",
       "      <td>and</td>\n",
       "      <td>CCONJ</td>\n",
       "      <td>757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4005</th>\n",
       "      <td>in</td>\n",
       "      <td>ADP</td>\n",
       "      <td>671</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6205</th>\n",
       "      <td>to</td>\n",
       "      <td>PART</td>\n",
       "      <td>660</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4743</th>\n",
       "      <td>on</td>\n",
       "      <td>ADP</td>\n",
       "      <td>482</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5567</th>\n",
       "      <td>said</td>\n",
       "      <td>VERB</td>\n",
       "      <td>451</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     token pos_tag  counts\n",
       "6145   the     DET    1903\n",
       "14       ,   PUNCT    1698\n",
       "21       .   PUNCT    1382\n",
       "4716    of     ADP     884\n",
       "1898     a     DET     789\n",
       "2093   and   CCONJ     757\n",
       "4005    in     ADP     671\n",
       "6205    to    PART     660\n",
       "4743    on     ADP     482\n",
       "5567  said    VERB     451"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# token frequency count (fact)\n",
    "pos_counts_fact = fact_tagsdf.groupby(['token','pos_tag']).size().reset_index(name='counts').sort_values(by='counts', ascending=False)\n",
    "pos_counts_fact.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pos_tag\n",
       "NOUN     2586\n",
       "VERB     1817\n",
       "PROPN    1672\n",
       "ADJ       882\n",
       "ADV       413\n",
       "NUM       221\n",
       "PRON       96\n",
       "ADP        89\n",
       "AUX        62\n",
       "SCONJ      51\n",
       "Name: token, dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# frequencies of pos tags\n",
    "pos_counts_fake.groupby(['pos_tag'])['token'].count().sort_values(ascending=False).head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pos_tag\n",
       "NOUN     2179\n",
       "VERB     1539\n",
       "PROPN    1379\n",
       "ADJ       747\n",
       "ADV       263\n",
       "NUM       205\n",
       "PRON       79\n",
       "ADP        70\n",
       "AUX        43\n",
       "SCONJ      42\n",
       "Name: token, dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos_counts_fact.groupby(['pos_tag'])['token'].count().sort_values(ascending=False).head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>token</th>\n",
       "      <th>pos_tag</th>\n",
       "      <th>counts</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5981</th>\n",
       "      <td>people</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>77</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7349</th>\n",
       "      <td>t</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6216</th>\n",
       "      <td>president</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7960</th>\n",
       "      <td>women</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7516</th>\n",
       "      <td>time</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3138</th>\n",
       "      <td>campaign</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8011</th>\n",
       "      <td>year</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4581</th>\n",
       "      <td>government</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5213</th>\n",
       "      <td>law</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8013</th>\n",
       "      <td>years</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7165</th>\n",
       "      <td>state</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4012</th>\n",
       "      <td>election</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3643</th>\n",
       "      <td>day</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5482</th>\n",
       "      <td>media</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3538</th>\n",
       "      <td>country</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>33</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           token pos_tag  counts\n",
       "5981      people    NOUN      77\n",
       "7349           t    NOUN      65\n",
       "6216   president    NOUN      58\n",
       "7960       women    NOUN      55\n",
       "7516        time    NOUN      52\n",
       "3138    campaign    NOUN      44\n",
       "8011        year    NOUN      44\n",
       "4581  government    NOUN      41\n",
       "5213         law    NOUN      40\n",
       "8013       years    NOUN      40\n",
       "7165       state    NOUN      38\n",
       "4012    election    NOUN      37\n",
       "3643         day    NOUN      35\n",
       "5482       media    NOUN      35\n",
       "3538     country    NOUN      33"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# dive into diferences in nouns\n",
    "pos_counts_fake[pos_counts_fake.pos_tag == \"NOUN\"][0:15]"
   ]
  },
  {
   "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>token</th>\n",
       "      <th>pos_tag</th>\n",
       "      <th>counts</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3738</th>\n",
       "      <td>government</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6618</th>\n",
       "      <td>year</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5901</th>\n",
       "      <td>state</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2360</th>\n",
       "      <td>bill</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1975</th>\n",
       "      <td>administration</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>51</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5066</th>\n",
       "      <td>president</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3277</th>\n",
       "      <td>election</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4919</th>\n",
       "      <td>people</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4786</th>\n",
       "      <td>order</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4259</th>\n",
       "      <td>law</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2497</th>\n",
       "      <td>campaign</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6095</th>\n",
       "      <td>tax</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5396</th>\n",
       "      <td>reporters</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5905</th>\n",
       "      <td>statement</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2873</th>\n",
       "      <td>court</td>\n",
       "      <td>NOUN</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               token pos_tag  counts\n",
       "3738      government    NOUN      71\n",
       "6618            year    NOUN      64\n",
       "5901           state    NOUN      57\n",
       "2360            bill    NOUN      55\n",
       "1975  administration    NOUN      51\n",
       "5066       president    NOUN      49\n",
       "3277        election    NOUN      48\n",
       "4919          people    NOUN      45\n",
       "4786           order    NOUN      45\n",
       "4259             law    NOUN      42\n",
       "2497        campaign    NOUN      42\n",
       "6095             tax    NOUN      39\n",
       "5396       reporters    NOUN      38\n",
       "5905       statement    NOUN      37\n",
       "2873           court    NOUN      37"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos_counts_fact[pos_counts_fact.pos_tag == \"NOUN\"][0:15]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Named Entities"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# top entities in fake news\n",
    "top_entities_fake = fake_tagsdf[fake_tagsdf['ner_tag'] != \"\"] \\\n",
    "                    .groupby(['token','ner_tag']).size().reset_index(name='counts') \\\n",
    "                    .sort_values(by='counts', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# top entities in fact news\n",
    "top_entities_fact = fact_tagsdf[fact_tagsdf['ner_tag'] != \"\"] \\\n",
    "                    .groupby(['token','ner_tag']).size().reset_index(name='counts') \\\n",
    "                    .sort_values(by='counts', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create custom palette to ensure plots are consistent\n",
    "ner_palette = {\n",
    "    'ORG': sns.color_palette(\"Set2\").as_hex()[0],\n",
    "    'GPE': sns.color_palette(\"Set2\").as_hex()[1],\n",
    "    'NORP': sns.color_palette(\"Set2\").as_hex()[2],\n",
    "    'PERSON': sns.color_palette(\"Set2\").as_hex()[3],\n",
    "    'DATE': sns.color_palette(\"Set2\").as_hex()[4],\n",
    "    'CARDINAL': sns.color_palette(\"Set2\").as_hex()[5],\n",
    "    'PERCENT': sns.color_palette(\"Set2\").as_hex()[6]\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Most Common Entities in Fake News')]"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.barplot(\n",
    "    x = 'counts',\n",
    "    y = 'token',\n",
    "    hue = 'ner_tag',\n",
    "    palette = ner_palette,\n",
    "    data = top_entities_fake[0:10],\n",
    "    orient = 'h',\n",
    "    dodge=False\n",
    ") \\\n",
    ".set(title='Most Common Entities in Fake News')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Most Common Entities in Factual News')]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.barplot(\n",
    "    x = 'counts',\n",
    "    y = 'token',\n",
    "    hue = 'ner_tag',\n",
    "    palette = ner_palette,\n",
    "    data = top_entities_fact[0:10],\n",
    "    orient = 'h',\n",
    "    dodge=False\n",
    ") \\\n",
    ".set(title='Most Common Entities in Factual News')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Text Pre-processing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "# a lot of the factual news has a location tag at the beginning of the article, let's use regex to remove this\n",
    "data['text_clean'] = data.apply(lambda x: re.sub(r\"^[^-]*-\\s*\", \"\", x['text']), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# lowercase \n",
    "data['text_clean'] = data['text_clean'].str.lower()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# remove punctuation\n",
    "data['text_clean'] = data.apply(lambda x: re.sub(r\"([^\\w\\s])\", \"\", x['text_clean']), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', \"you're\", \"you've\", \"you'll\", \"you'd\", 'your', 'yours', 'yourself', 'yourselves', 'he', 'him', 'his', 'himself', 'she', \"she's\", 'her', 'hers', 'herself', 'it', \"it's\", 'its', 'itself', 'they', 'them', 'their', 'theirs', 'themselves', 'what', 'which', 'who', 'whom', 'this', 'that', \"that'll\", 'these', 'those', 'am', 'is', 'are', 'was', 'were', 'be', 'been', 'being', 'have', 'has', 'had', 'having', 'do', 'does', 'did', 'doing', 'a', 'an', 'the', 'and', 'but', 'if', 'or', 'because', 'as', 'until', 'while', 'of', 'at', 'by', 'for', 'with', 'about', 'against', 'between', 'into', 'through', 'during', 'before', 'after', 'above', 'below', 'to', 'from', 'up', 'down', 'in', 'out', 'on', 'off', 'over', 'under', 'again', 'further', 'then', 'once', 'here', 'there', 'when', 'where', 'why', 'how', 'all', 'any', 'both', 'each', 'few', 'more', 'most', 'other', 'some', 'such', 'no', 'nor', 'not', 'only', 'own', 'same', 'so', 'than', 'too', 'very', 's', 't', 'can', 'will', 'just', 'don', \"don't\", 'should', \"should've\", 'now', 'd', 'll', 'm', 'o', 're', 've', 'y', 'ain', 'aren', \"aren't\", 'couldn', \"couldn't\", 'didn', \"didn't\", 'doesn', \"doesn't\", 'hadn', \"hadn't\", 'hasn', \"hasn't\", 'haven', \"haven't\", 'isn', \"isn't\", 'ma', 'mightn', \"mightn't\", 'mustn', \"mustn't\", 'needn', \"needn't\", 'shan', \"shan't\", 'shouldn', \"shouldn't\", 'wasn', \"wasn't\", 'weren', \"weren't\", 'won', \"won't\", 'wouldn', \"wouldn't\"]\n"
     ]
    }
   ],
   "source": [
    "# stop words\n",
    "en_stopwords = stopwords.words('english')\n",
    "print(en_stopwords) # check this against our most frequent n-grams"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['text_clean'] = data['text_clean'].apply(lambda x: ' '.join([word for word in x.split() if word not in (en_stopwords)]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "# tokenize \n",
    "data['text_clean'] = data.apply(lambda x: word_tokenize(x['text_clean']), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# lemmatize\n",
    "lemmatizer = WordNetLemmatizer()\n",
    "data[\"text_clean\"] = data[\"text_clean\"].apply(lambda tokens: [lemmatizer.lemmatize(token) for token in tokens])"
   ]
  },
  {
   "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>title</th>\n",
       "      <th>text</th>\n",
       "      <th>date</th>\n",
       "      <th>fake_or_factual</th>\n",
       "      <th>text_clean</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>HOLLYWEIRD LIB SUSAN SARANDON Compares Muslim ...</td>\n",
       "      <td>There are two small problems with your analogy...</td>\n",
       "      <td>Dec 30, 2015</td>\n",
       "      <td>Fake News</td>\n",
       "      <td>[yearold, oscarwinning, actress, described, me...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Elijah Cummings Called Trump Out To His Face ...</td>\n",
       "      <td>Buried in Trump s bonkers interview with New Y...</td>\n",
       "      <td>April 6, 2017</td>\n",
       "      <td>Fake News</td>\n",
       "      <td>[buried, trump, bonkers, interview, new, york,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Hillary Clinton Says Half Her Cabinet Will Be...</td>\n",
       "      <td>Women make up over 50 percent of this country,...</td>\n",
       "      <td>April 26, 2016</td>\n",
       "      <td>Fake News</td>\n",
       "      <td>[woman, make, 50, percent, country, grossly, u...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Russian bombing of U.S.-backed forces being di...</td>\n",
       "      <td>WASHINGTON (Reuters) - U.S. Defense Secretary ...</td>\n",
       "      <td>September 18, 2017</td>\n",
       "      <td>Factual News</td>\n",
       "      <td>[u, defense, secretary, jim, mattis, said, mon...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Britain says window to restore Northern Irelan...</td>\n",
       "      <td>BELFAST (Reuters) - Northern Ireland s politic...</td>\n",
       "      <td>September 4, 2017</td>\n",
       "      <td>Factual News</td>\n",
       "      <td>[northern, ireland, political, party, rapidly,...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               title  \\\n",
       "0  HOLLYWEIRD LIB SUSAN SARANDON Compares Muslim ...   \n",
       "1   Elijah Cummings Called Trump Out To His Face ...   \n",
       "2   Hillary Clinton Says Half Her Cabinet Will Be...   \n",
       "3  Russian bombing of U.S.-backed forces being di...   \n",
       "4  Britain says window to restore Northern Irelan...   \n",
       "\n",
       "                                                text                 date  \\\n",
       "0  There are two small problems with your analogy...         Dec 30, 2015   \n",
       "1  Buried in Trump s bonkers interview with New Y...        April 6, 2017   \n",
       "2  Women make up over 50 percent of this country,...       April 26, 2016   \n",
       "3  WASHINGTON (Reuters) - U.S. Defense Secretary ...  September 18, 2017    \n",
       "4  BELFAST (Reuters) - Northern Ireland s politic...   September 4, 2017    \n",
       "\n",
       "  fake_or_factual                                         text_clean  \n",
       "0       Fake News  [yearold, oscarwinning, actress, described, me...  \n",
       "1       Fake News  [buried, trump, bonkers, interview, new, york,...  \n",
       "2       Fake News  [woman, make, 50, percent, country, grossly, u...  \n",
       "3    Factual News  [u, defense, secretary, jim, mattis, said, mon...  \n",
       "4    Factual News  [northern, ireland, political, party, rapidly,...  "
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "           index  count\n",
      "0        (said,)    560\n",
      "1       (trump,)    520\n",
      "2           (u,)    255\n",
      "3       (state,)    250\n",
      "4   (president,)    226\n",
      "5       (would,)    210\n",
      "6         (one,)    141\n",
      "7        (year,)    128\n",
      "8  (republican,)    128\n",
      "9        (also,)    124\n"
     ]
    }
   ],
   "source": [
    "# most common unigrams after preprocessing\n",
    "tokens_clean = sum(data['text_clean'], [])\n",
    "unigrams = (pd.Series(nltk.ngrams(tokens_clean, 1)).value_counts()).reset_index()[:10]\n",
    "print(unigrams)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\lnewb\\AppData\\Local\\Temp\\ipykernel_14668\\4019870307.py:3: UserWarning: \n",
      "The palette list has fewer values (1) than needed (10) and will cycle, which may produce an uninterpretable plot.\n",
      "  sns.barplot(x = \"count\",\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Most Common Unigrams After Preprocessing')]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "unigrams['token'] = unigrams['index'].apply(lambda x: x[0]) # extract the token from the tuple so we can plot it\n",
    "\n",
    "sns.barplot(x = \"count\", \n",
    "            y = \"token\", \n",
    "            data=unigrams,\n",
    "            orient = 'h',\n",
    "            palette=[default_plot_colour],\n",
    "            hue = \"token\", legend = False)\\\n",
    ".set(title='Most Common Unigrams After Preprocessing')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(donald, trump)        92\n",
      "(united, state)        80\n",
      "(white, house)         72\n",
      "(president, donald)    42\n",
      "(hillary, clinton)     31\n",
      "(new, york)            31\n",
      "(image, via)           29\n",
      "(supreme, court)       29\n",
      "(official, said)       26\n",
      "(food, stamp)          24\n",
      "Name: count, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# most common bigrams after preprocessing\n",
    "bigrams = (pd.Series(nltk.ngrams(tokens_clean, 2)).value_counts()) \n",
    "print(bigrams[:10])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Sentiment Analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "# use vader so we also get a neutral sentiment count\n",
    "vader_sentiment = SentimentIntensityAnalyzer()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['vader_sentiment_score'] = data['text'].apply(lambda review: vader_sentiment.polarity_scores(review)['compound'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "# create labels\n",
    "bins = [-1, -0.1, 0.1, 1]\n",
    "names = ['negative', 'neutral', 'positive']\n",
    "\n",
    "data['vader_sentiment_label'] = pd.cut(data['vader_sentiment_score'], bins, labels=names)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='vader_sentiment_label'>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "data['vader_sentiment_label'].value_counts().plot.bar(color=default_plot_colour)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\lnewb\\AppData\\Local\\Temp\\ipykernel_14668\\3014756626.py:1: UserWarning: The palette list has more values (6) than needed (3), which may not be intended.\n",
      "  sns.countplot(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 1.0, 'Sentiment by News Type')]"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.countplot(\n",
    "    x = 'fake_or_factual',\n",
    "    hue = 'vader_sentiment_label',\n",
    "    palette = sns.color_palette(\"hls\"),\n",
    "    data = data\n",
    ") \\\n",
    ".set(title='Sentiment by News Type')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## LDA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "# fake news data vectorization\n",
    "fake_news_text = data[data['fake_or_factual'] == \"Fake News\"]['text_clean'].reset_index(drop=True)\n",
    "dictionary_fake = corpora.Dictionary(fake_news_text)\n",
    "doc_term_fake = [dictionary_fake.doc2bow(text) for text in fake_news_text]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# generate coherence scores to determine an optimum number of topics\n",
    "coherence_values = []\n",
    "model_list = []\n",
    "\n",
    "min_topics = 2\n",
    "max_topics = 11\n",
    "\n",
    "for num_topics_i in range(min_topics, max_topics+1):\n",
    "    model = gensim.models.LdaModel(doc_term_fake, num_topics=num_topics_i, id2word = dictionary_fake)\n",
    "    model_list.append(model)\n",
    "    coherence_model = CoherenceModel(model=model, texts=fake_news_text, dictionary=dictionary_fake, coherence='c_v')\n",
    "    coherence_values.append(coherence_model.get_coherence())\n",
    "    \n",
    "plt.plot(range(min_topics, max_topics+1), coherence_values)\n",
    "plt.xlabel(\"Number of Topics\")\n",
    "plt.ylabel(\"Coherence score\")\n",
    "plt.legend((\"coherence_values\"), loc='best')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(0,\n",
       "  '0.009*\"trump\" + 0.004*\"state\" + 0.004*\"president\" + 0.004*\"clinton\" + 0.004*\"school\" + 0.004*\"said\" + 0.003*\"one\" + 0.003*\"people\" + 0.003*\"official\" + 0.003*\"go\"'),\n",
       " (1,\n",
       "  '0.009*\"trump\" + 0.004*\"republican\" + 0.003*\"would\" + 0.003*\"state\" + 0.003*\"said\" + 0.003*\"time\" + 0.003*\"u\" + 0.002*\"test\" + 0.002*\"new\" + 0.002*\"like\"'),\n",
       " (2,\n",
       "  '0.011*\"trump\" + 0.005*\"said\" + 0.005*\"president\" + 0.005*\"one\" + 0.005*\"u\" + 0.004*\"time\" + 0.004*\"would\" + 0.004*\"donald\" + 0.003*\"clinton\" + 0.003*\"say\"'),\n",
       " (3,\n",
       "  '0.016*\"trump\" + 0.005*\"said\" + 0.005*\"republican\" + 0.005*\"clinton\" + 0.003*\"time\" + 0.003*\"state\" + 0.003*\"woman\" + 0.003*\"one\" + 0.003*\"student\" + 0.003*\"would\"'),\n",
       " (4,\n",
       "  '0.010*\"trump\" + 0.006*\"clinton\" + 0.004*\"u\" + 0.004*\"obama\" + 0.004*\"president\" + 0.004*\"state\" + 0.003*\"would\" + 0.003*\"said\" + 0.003*\"people\" + 0.003*\"republican\"'),\n",
       " (5,\n",
       "  '0.009*\"trump\" + 0.007*\"said\" + 0.005*\"would\" + 0.005*\"president\" + 0.004*\"state\" + 0.004*\"people\" + 0.004*\"clinton\" + 0.003*\"time\" + 0.003*\"year\" + 0.003*\"one\"')]"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create lda model\n",
    "num_topics_fake = 6 \n",
    "\n",
    "lda_model_fake = gensim.models.LdaModel(corpus=doc_term_fake,\n",
    "                                       id2word=dictionary_fake,\n",
    "                                       num_topics=num_topics_fake)\n",
    "\n",
    "lda_model_fake.print_topics(num_topics=num_topics_fake, num_words=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "# our topics contain a lot of very similar words, let's try using latent semantic anaysis with tf-idf vectorization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## TF-IDF & LSA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "def tfidf_corpus(doc_term_matrix):\n",
    "    # create a corpus using tfidf vecotization\n",
    "    tfidf = TfidfModel(corpus=doc_term_matrix, normalize=True)\n",
    "    corpus_tfidf = tfidf[doc_term_matrix]\n",
    "    return corpus_tfidf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_coherence_scores(corpus, dictionary, text, min_topics, max_topics):\n",
    "    # generate coherence scores to determine an optimum number of topics\n",
    "    coherence_values = []\n",
    "    model_list = []\n",
    "    for num_topics_i in range(min_topics, max_topics+1):\n",
    "        model = LsiModel(corpus, num_topics=num_topics_i, id2word = dictionary, random_seed=0)\n",
    "        model_list.append(model)\n",
    "        coherence_model = CoherenceModel(model=model, texts=text, dictionary=dictionary, coherence='c_v')\n",
    "        coherence_values.append(coherence_model.get_coherence())\n",
    "    # plot results\n",
    "    plt.plot(range(min_topics, max_topics+1), coherence_values)\n",
    "    plt.xlabel(\"Number of Topics\")\n",
    "    plt.ylabel(\"Coherence score\")\n",
    "    plt.legend((\"coherence_values\"), loc='best')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# create tfidf representation\n",
    "corpus_tfidf_fake = tfidf_corpus(doc_term_fake)\n",
    "# coherence scores for fake news data\n",
    "get_coherence_scores(corpus_tfidf_fake, dictionary_fake, fake_news_text, min_topics=2, max_topics=11)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(0,\n",
       "  '0.218*\"trump\" + 0.135*\"clinton\" + 0.094*\"woman\" + 0.087*\"president\" + 0.086*\"republican\" + 0.085*\"obama\" + 0.084*\"party\" + 0.083*\"school\" + 0.081*\"said\" + 0.079*\"time\"'),\n",
       " (1,\n",
       "  '0.299*\"boiler\" + 0.253*\"room\" + 0.250*\"acr\" + 0.186*\"jay\" + 0.185*\"animal\" + 0.176*\"episode\" + 0.147*\"analysis\" + 0.122*\"dyer\" + 0.119*\"savant\" + 0.119*\"moralist\"'),\n",
       " (2,\n",
       "  '-0.219*\"school\" + 0.195*\"clinton\" + 0.163*\"conference\" + -0.151*\"county\" + -0.137*\"student\" + 0.120*\"press\" + 0.116*\"trump\" + 0.112*\"hillary\" + -0.101*\"love\" + 0.095*\"email\"'),\n",
       " (3,\n",
       "  '0.375*\"flynn\" + 0.151*\"russian\" + 0.111*\"email\" + 0.109*\"30\" + 0.101*\"department\" + 0.100*\"information\" + 0.100*\"nana\" + 0.099*\"official\" + 0.098*\"mccain\" + -0.096*\"trump\"'),\n",
       " (4,\n",
       "  '-0.235*\"trump\" + -0.223*\"flynn\" + -0.166*\"conference\" + 0.165*\"clinton\" + 0.153*\"dnc\" + 0.151*\"sander\" + 0.127*\"email\" + -0.111*\"press\" + 0.109*\"rich\" + -0.084*\"tax\"')]"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# model for fake news data\n",
    "lsa_fake = LsiModel(corpus_tfidf_fake, id2word=dictionary_fake, num_topics=5)\n",
    "lsa_fake.print_topics()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predict fake or factual news "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "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>title</th>\n",
       "      <th>text</th>\n",
       "      <th>date</th>\n",
       "      <th>fake_or_factual</th>\n",
       "      <th>text_clean</th>\n",
       "      <th>vader_sentiment_score</th>\n",
       "      <th>vader_sentiment_label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>HOLLYWEIRD LIB SUSAN SARANDON Compares Muslim ...</td>\n",
       "      <td>There are two small problems with your analogy...</td>\n",
       "      <td>Dec 30, 2015</td>\n",
       "      <td>Fake News</td>\n",
       "      <td>[yearold, oscarwinning, actress, described, me...</td>\n",
       "      <td>-0.3660</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Elijah Cummings Called Trump Out To His Face ...</td>\n",
       "      <td>Buried in Trump s bonkers interview with New Y...</td>\n",
       "      <td>April 6, 2017</td>\n",
       "      <td>Fake News</td>\n",
       "      <td>[buried, trump, bonkers, interview, new, york,...</td>\n",
       "      <td>-0.8197</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Hillary Clinton Says Half Her Cabinet Will Be...</td>\n",
       "      <td>Women make up over 50 percent of this country,...</td>\n",
       "      <td>April 26, 2016</td>\n",
       "      <td>Fake News</td>\n",
       "      <td>[woman, make, 50, percent, country, grossly, u...</td>\n",
       "      <td>0.9779</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Russian bombing of U.S.-backed forces being di...</td>\n",
       "      <td>WASHINGTON (Reuters) - U.S. Defense Secretary ...</td>\n",
       "      <td>September 18, 2017</td>\n",
       "      <td>Factual News</td>\n",
       "      <td>[u, defense, secretary, jim, mattis, said, mon...</td>\n",
       "      <td>-0.3400</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Britain says window to restore Northern Irelan...</td>\n",
       "      <td>BELFAST (Reuters) - Northern Ireland s politic...</td>\n",
       "      <td>September 4, 2017</td>\n",
       "      <td>Factual News</td>\n",
       "      <td>[northern, ireland, political, party, rapidly,...</td>\n",
       "      <td>0.8590</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               title  \\\n",
       "0  HOLLYWEIRD LIB SUSAN SARANDON Compares Muslim ...   \n",
       "1   Elijah Cummings Called Trump Out To His Face ...   \n",
       "2   Hillary Clinton Says Half Her Cabinet Will Be...   \n",
       "3  Russian bombing of U.S.-backed forces being di...   \n",
       "4  Britain says window to restore Northern Irelan...   \n",
       "\n",
       "                                                text                 date  \\\n",
       "0  There are two small problems with your analogy...         Dec 30, 2015   \n",
       "1  Buried in Trump s bonkers interview with New Y...        April 6, 2017   \n",
       "2  Women make up over 50 percent of this country,...       April 26, 2016   \n",
       "3  WASHINGTON (Reuters) - U.S. Defense Secretary ...  September 18, 2017    \n",
       "4  BELFAST (Reuters) - Northern Ireland s politic...   September 4, 2017    \n",
       "\n",
       "  fake_or_factual                                         text_clean  \\\n",
       "0       Fake News  [yearold, oscarwinning, actress, described, me...   \n",
       "1       Fake News  [buried, trump, bonkers, interview, new, york,...   \n",
       "2       Fake News  [woman, make, 50, percent, country, grossly, u...   \n",
       "3    Factual News  [u, defense, secretary, jim, mattis, said, mon...   \n",
       "4    Factual News  [northern, ireland, political, party, rapidly,...   \n",
       "\n",
       "   vader_sentiment_score vader_sentiment_label  \n",
       "0                -0.3660              negative  \n",
       "1                -0.8197              negative  \n",
       "2                 0.9779              positive  \n",
       "3                -0.3400              negative  \n",
       "4                 0.8590              positive  "
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = [','.join(map(str, l)) for l in data['text_clean']]\n",
    "Y = data['fake_or_factual']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "# text vectorization - CountVectorizer\n",
    "countvec = CountVectorizer()\n",
    "countvec_fit = countvec.fit_transform(X)\n",
    "bag_of_words = pd.DataFrame(countvec_fit.toarray(), columns = countvec.get_feature_names_out())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "# split into train and test data\n",
    "X_train, X_test, y_train, y_test = train_test_split(bag_of_words, Y, test_size=0.3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "lr = LogisticRegression(random_state=0).fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_lr = lr.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8166666666666667"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(y_pred_lr, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "Factual News       0.78      0.86      0.82        29\n",
      "   Fake News       0.86      0.77      0.81        31\n",
      "\n",
      "    accuracy                           0.82        60\n",
      "   macro avg       0.82      0.82      0.82        60\n",
      "weighted avg       0.82      0.82      0.82        60\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, y_pred_lr))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "svm = SGDClassifier().fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_svm = svm.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8666666666666667"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accuracy_score(y_pred_svm, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "Factual News       0.86      0.86      0.86        29\n",
      "   Fake News       0.87      0.87      0.87        31\n",
      "\n",
      "    accuracy                           0.87        60\n",
      "   macro avg       0.87      0.87      0.87        60\n",
      "weighted avg       0.87      0.87      0.87        60\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_test, y_pred_svm))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "nlp_course_env",
   "language": "python",
   "name": "nlp_course_env"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}