Training.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing Important libraries\n",
    "\n",
    "import re\n",
    "import pickle\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.stem import PorterStemmer\n",
    "from matplotlib import pyplot as plt\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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>index</th>\n",
       "      <th>title</th>\n",
       "      <th>text</th>\n",
       "      <th>subject</th>\n",
       "      <th>date</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>22216</td>\n",
       "      <td>Ben Stein Calls Out 9th Circuit Court: Committ...</td>\n",
       "      <td>21st Century Wire says Ben Stein, reputable pr...</td>\n",
       "      <td>US_News</td>\n",
       "      <td>February 13, 2017</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4436</td>\n",
       "      <td>Trump drops Steve Bannon from National Securit...</td>\n",
       "      <td>WASHINGTON (Reuters) - U.S. President Donald T...</td>\n",
       "      <td>politicsNews</td>\n",
       "      <td>April 5, 2017</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1526</td>\n",
       "      <td>Puerto Rico expects U.S. to lift Jones Act shi...</td>\n",
       "      <td>(Reuters) - Puerto Rico Governor Ricardo Rosse...</td>\n",
       "      <td>politicsNews</td>\n",
       "      <td>September 27, 2017</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1377</td>\n",
       "      <td>OOPS: Trump Just Accidentally Confirmed He Le...</td>\n",
       "      <td>On Monday, Donald Trump once again embarrassed...</td>\n",
       "      <td>News</td>\n",
       "      <td>May 22, 2017</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>8995</td>\n",
       "      <td>Donald Trump heads for Scotland to reopen a go...</td>\n",
       "      <td>GLASGOW, Scotland (Reuters) - Most U.S. presid...</td>\n",
       "      <td>politicsNews</td>\n",
       "      <td>June 24, 2016</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   index                                              title  \\\n",
       "0  22216  Ben Stein Calls Out 9th Circuit Court: Committ...   \n",
       "1   4436  Trump drops Steve Bannon from National Securit...   \n",
       "2   1526  Puerto Rico expects U.S. to lift Jones Act shi...   \n",
       "3   1377   OOPS: Trump Just Accidentally Confirmed He Le...   \n",
       "4   8995  Donald Trump heads for Scotland to reopen a go...   \n",
       "\n",
       "                                                text       subject  \\\n",
       "0  21st Century Wire says Ben Stein, reputable pr...       US_News   \n",
       "1  WASHINGTON (Reuters) - U.S. President Donald T...  politicsNews   \n",
       "2  (Reuters) - Puerto Rico Governor Ricardo Rosse...  politicsNews   \n",
       "3  On Monday, Donald Trump once again embarrassed...          News   \n",
       "4  GLASGOW, Scotland (Reuters) - Most U.S. presid...  politicsNews   \n",
       "\n",
       "                  date  class  \n",
       "0    February 13, 2017      0  \n",
       "1       April 5, 2017       1  \n",
       "2  September 27, 2017       1  \n",
       "3         May 22, 2017      0  \n",
       "4       June 24, 2016       1  "
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Reading in both real and fake data\n",
    "fake = pd.read_csv(\"Dataset/Fake.csv\")\n",
    "real = pd.read_csv(\"Dataset/True.csv\")\n",
    "\n",
    "# Adding class column to fake and real data sets.\n",
    "# 0 for fake\n",
    "# 1 for real\n",
    "fake[\"class\"] = 0\n",
    "real[\"class\"] = 1\n",
    "\n",
    "# Concatenating datasets\n",
    "data = pd.concat([fake, real], axis=0)\n",
    "# Shuffling data\n",
    "data = data.sample(frac= 1.0, random_state= 42).reset_index()\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Saving combined data\n",
    "# data.to_csv(\"Dataset\\Combined.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dropping unnecessary columns\n",
    "data.drop([\"index\", \"title\", \"subject\", \"date\"], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 44898 entries, 0 to 44897\n",
      "Data columns (total 2 columns):\n",
      " #   Column  Non-Null Count  Dtype \n",
      "---  ------  --------------  ----- \n",
      " 0   text    44898 non-null  object\n",
      " 1   class   44898 non-null  int64 \n",
      "dtypes: int64(1), object(1)\n",
      "memory usage: 701.7+ KB\n"
     ]
    }
   ],
   "source": [
    "# Basic information about the data\n",
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "text     0\n",
       "class    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Number of empty rows on each column\n",
    "\n",
    "data.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "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>text</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>21st Century Wire says Ben Stein, reputable pr...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>WASHINGTON (Reuters) - U.S. President Donald T...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(Reuters) - Puerto Rico Governor Ricardo Rosse...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>On Monday, Donald Trump once again embarrassed...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>GLASGOW, Scotland (Reuters) - Most U.S. presid...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text  class\n",
       "0  21st Century Wire says Ben Stein, reputable pr...      0\n",
       "1  WASHINGTON (Reuters) - U.S. President Donald T...      1\n",
       "2  (Reuters) - Puerto Rico Governor Ricardo Rosse...      1\n",
       "3  On Monday, Donald Trump once again embarrassed...      0\n",
       "4  GLASGOW, Scotland (Reuters) - Most U.S. presid...      1"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "counts = data[\"class\"].value_counts()\n",
    "\n",
    "plt.bar([\"Fake_news\", \"Real_news\"], counts.values)\n",
    "plt.title(\"News Type Amount\", fontsize= 20)\n",
    "plt.xlabel(\"News Type\", fontsize=15)\n",
    "plt.ylabel(\"Amount\", fontsize=15)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fake News for obeservation - Text\n",
      "21st Century Wire says Ben Stein, reputable professor from, Pepperdine University (also of some Hollywood fame appearing in TV shows and films such as Ferris Bueller s Day Off) made some provocative statements on Judge Jeanine Pirro s show recently. While discussing the halt that was imposed on President Trump s Executive Order on travel. Stein referred to the judgement by the 9th Circuit Court in Washington state as a  Coup d tat against the executive branch and against the constitution.  Stein went on to call the Judges in Seattle  political puppets  and the judiciary  political pawns. Watch the interview below for the complete statements and note the stark contrast to the rhetoric of the leftist media and pundits who neglect to note that no court has ever blocked any Presidential orders in immigration in the past or discuss the legal efficacy of the halt or the actual text of the Executive Order.READ MORE TRUMP NEWS AT: 21st Century Wire Trump FilesSUPPORT OUR WORK BY SUBSCRIBING & BECOMING A MEMBER @21WIRE.TV\n",
      "0\n",
      "\n",
      "Real News for observation\n",
      "WASHINGTON (Reuters) - U.S. President Donald Trump removed his chief strategist Steve Bannon from the National Security Council on Wednesday, reversing his controversial decision early this year to give a political adviser an unprecedented role in security discussions. Trump’s overhaul of the NSC, confirmed by a White House official, also elevated General Joseph Dunford, chairman of the Joint Chiefs of Staff and Dan Coats, the director of National Intelligence who heads all 17 U.S. intelligence agencies. The official said the change moves the NSC “back to its core function of what it’s supposed to do.” It also appears to mark a victory for national security adviser H.R. McMaster, who had told some national security experts he felt he was in a “battle to the death” with Bannon and others on the White House staff. Vice President Mike Pence said Bannon would continue to play an important role in policy and played down the shake-up as routine. “This is just a natural evolution to ensure the National Security Council is organized in a way that best serves the president in resolving and making those difficult decisions,” Pence said on Fox News.  Bannon said in a statement he had succeeded in returning the NSC back to its traditional role of coordinating foreign policy rather than running it. He cited President Barack Obama’s national security adviser, Susan Rice, for why he advocated a change. “Susan Rice operationalized the NSC during the last administration so I was put on NSC to ensure it was ‘de-operationalized.’ General McMaster has NSC back to its proper function,” he said. Trump’s White House team has grappled with infighting and intrigue that has hobbled his young presidency. In recent days, several other senior U.S. foreign policy and national security officials have said the mechanisms for shaping the Trump administration’s response to pressing challenges such as Syria, North Korea and Iran still were not in place.  Critics of Bannon’s role on the NSC said it gave too much weight in decision-making to someone who lacked foreign policy expertise. Bannon, who was chief executive of Trump’s presidential campaign in the months leading to his election in November, in some respects represents Trump’s “America First” nationalistic voice, helping fuel his anti-Washington fervor and pushing for the president to part ways at times with mainstream Republicans. Before joining the Trump administration, Bannon headed Breitbart News, a right-wing website. U.S. Representative Adam Schiff, ranking Democrat on the House of Representatives Intelligence Committee, called the shift in the NSC a positive step that will help McMaster “gain control over a body that was being politicized by Bannon’s involvement.” “As the administration’s policy over North Korea, China, Russia and Syria continues to drift, we can only hope this shake-up brings some level of strategic vision to the body,” he said. Bannon’s removal from the NSC was a potential setback for his sphere of influence in the Trump White House, where he has a voice in most major decisions. But a Trump confidant said Bannon remained as influential as ever. “He is still involved in everything and still has the full confidence of the president but to be fair he can only do so much stuff,” the confidant said, speaking on condition of anonymity. The White House official said Bannon was no longer needed on the NSC after the departure of Trump’s first national security adviser, Michael Flynn. Flynn was forced to resign on Feb. 13 over his contacts with Russia’s ambassador to the United States, Sergei Kislyak, prior to Trump taking office on Jan. 20. The official said Bannon had been placed on the NSC originally as a check on Flynn and had only attended one of the NSC’s regular meetings. The official dismissed questions about a power struggle between Bannon and McMaster, saying they shared the same world view. However, two current national security officials rejected the White House explanation, noting that two months have passed since Flynn’s departure. McMaster, they said, speaking on the condition of anonymity, also has dueled with Bannon and others over direct access to Trump; the future of deputy national security adviser K.T. McFarland, a former Fox News commentator; intelligence director Ezra Cohen-Watnick, a Flynn appointee; and other staffing decisions. Trump is preparing for his first face-to-face meeting on Thursday and Friday with Chinese President Xi Jinping with the threat of North Korea’s nuclear and missile programs a key component of their talks. Bannon’s seat on the NSC’s “principals’ committee,” a group that includes the secretaries of state, defense and other ranking aides, was taken by Rick Perry, who as energy secretary is charged with overseeing the U.S. nuclear weapons arsenal. \n",
      "1\n"
     ]
    }
   ],
   "source": [
    "# Data cleaning/features engineering.\n",
    "\n",
    "print(\"Fake News for obeservation - Text\")\n",
    "print(data.text[0])\n",
    "print(data[\"class\"][0])\n",
    "print(\"\\nReal News for observation\")\n",
    "print(data.text[1])\n",
    "print(data[\"class\"][1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "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>text</th>\n",
       "      <th>class</th>\n",
       "      <th>len_with_stpwds</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>21st Century Wire says Ben Stein, reputable pr...</td>\n",
       "      <td>0</td>\n",
       "      <td>1028</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>WASHINGTON (Reuters) - U.S. President Donald T...</td>\n",
       "      <td>1</td>\n",
       "      <td>4820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(Reuters) - Puerto Rico Governor Ricardo Rosse...</td>\n",
       "      <td>1</td>\n",
       "      <td>1848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>On Monday, Donald Trump once again embarrassed...</td>\n",
       "      <td>0</td>\n",
       "      <td>1244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>GLASGOW, Scotland (Reuters) - Most U.S. presid...</td>\n",
       "      <td>1</td>\n",
       "      <td>3137</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text  class  len_with_stpwds\n",
       "0  21st Century Wire says Ben Stein, reputable pr...      0             1028\n",
       "1  WASHINGTON (Reuters) - U.S. President Donald T...      1             4820\n",
       "2  (Reuters) - Puerto Rico Governor Ricardo Rosse...      1             1848\n",
       "3  On Monday, Donald Trump once again embarrassed...      0             1244\n",
       "4  GLASGOW, Scotland (Reuters) - Most U.S. presid...      1             3137"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Adding length of news before removing non-essentials.\n",
    "data[\"len_with_stpwds\"] = data[\"text\"].apply(lambda x: len(x))\n",
    "\n",
    "# Viewing data with text lengths\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "class\n",
      "0    2547.396235\n",
      "1    2383.278517\n",
      "Name: len_with_stpwds, dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Checking if text length has any effect on the probability of a news piece being true or false\n",
    "\n",
    "len_class = data.groupby(\"class\")[\"len_with_stpwds\"].mean()\n",
    "print(len_class)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> Based on the result of the analysis, there is no real difference between the length of fake or real news therefore, the column will be dropped."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(\"len_with_stpwds\", axis= 1, inplace = True)"
   ]
  },
  {
   "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>text</th>\n",
       "      <th>class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>21st Century Wire says Ben Stein, reputable pr...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>WASHINGTON (Reuters) - U.S. President Donald T...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(Reuters) - Puerto Rico Governor Ricardo Rosse...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>On Monday, Donald Trump once again embarrassed...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>GLASGOW, Scotland (Reuters) - Most U.S. presid...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text  class\n",
       "0  21st Century Wire says Ben Stein, reputable pr...      0\n",
       "1  WASHINGTON (Reuters) - U.S. President Donald T...      1\n",
       "2  (Reuters) - Puerto Rico Governor Ricardo Rosse...      1\n",
       "3  On Monday, Donald Trump once again embarrassed...      0\n",
       "4  GLASGOW, Scotland (Reuters) - Most U.S. presid...      1"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "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>text</th>\n",
       "      <th>class</th>\n",
       "      <th>formatted_text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>21st Century Wire says Ben Stein, reputable pr...</td>\n",
       "      <td>0</td>\n",
       "      <td>st century wire says ben stein reputable profe...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>WASHINGTON (Reuters) - U.S. President Donald T...</td>\n",
       "      <td>1</td>\n",
       "      <td>washington reuters  us president donald trump ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(Reuters) - Puerto Rico Governor Ricardo Rosse...</td>\n",
       "      <td>1</td>\n",
       "      <td>reuters  puerto rico governor ricardo rossell...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>On Monday, Donald Trump once again embarrassed...</td>\n",
       "      <td>0</td>\n",
       "      <td>on monday donald trump once again embarrassed ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>GLASGOW, Scotland (Reuters) - Most U.S. presid...</td>\n",
       "      <td>1</td>\n",
       "      <td>glasgow scotland reuters  most us presidential...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text  class  \\\n",
       "0  21st Century Wire says Ben Stein, reputable pr...      0   \n",
       "1  WASHINGTON (Reuters) - U.S. President Donald T...      1   \n",
       "2  (Reuters) - Puerto Rico Governor Ricardo Rosse...      1   \n",
       "3  On Monday, Donald Trump once again embarrassed...      0   \n",
       "4  GLASGOW, Scotland (Reuters) - Most U.S. presid...      1   \n",
       "\n",
       "                                      formatted_text  \n",
       "0  st century wire says ben stein reputable profe...  \n",
       "1  washington reuters  us president donald trump ...  \n",
       "2   reuters  puerto rico governor ricardo rossell...  \n",
       "3  on monday donald trump once again embarrassed ...  \n",
       "4  glasgow scotland reuters  most us presidential...  "
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Text formatting: lowering text cases, removal of punctuations, removal of nubers.\n",
    "\n",
    "# Turning text to lower case\n",
    "data[\"formatted_text\"] = data[\"text\"].apply(lambda x: x.lower())\n",
    "\n",
    "# Removal of numnbers\n",
    "data[\"formatted_text\"] = data[\"formatted_text\"].apply(lambda x: re.sub(r'[0-9]+', \"\", x))\n",
    "\n",
    "# Removal of punctuations, symbols and whitespaces\n",
    "data[\"formatted_text\"] = data[\"formatted_text\"].apply(lambda x: x.replace(\",\", \" \").replace(\":\", \" \").replace(\".\", \"\")\n",
    ".replace(\"'\", \" \").replace(\"%\", \" \").replace(\";\", \" \").replace(\"/\", \" \").replace('\"', \" \").replace(\"(\", \" \")\n",
    ".replace(\")\", \" \").replace(\"!\", \" \").replace(\"+\", \" \").replace(\"@\", \" \").replace(\"*\", \" \").replace(\"$\", \" \")\n",
    ".replace(\"&\", \" \").replace(\"-\", \" \").replace(\"?\", \" \").replace(\"#\", \" \").replace(\"[\", \" \").replace(\"]\", \" \")\n",
    ".replace(\"“\", \" \").replace(\"”\", \" \").replace(r\"‘\", \" \").replace(r\"’\", \" \").replace(\"'\", \" \")\n",
    ".replace(\"  \", \" \").replace(\"    \", \" \").replace(\"_\", \" \").replace(\"\\n\", \" \").replace(\"`\", \" \").replace(\"\\t\", \" \"))\n",
    "\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Instantiation of stemmer and english stopwords\n",
    "eng_stop_words = stopwords.words(\"english\")\n",
    "stemmer = PorterStemmer()\n",
    "\n",
    "\n",
    "# Removal of stopwords, stemming and removal of words with insignificant length\n",
    "data[\"formatted_text\"] = data[\"formatted_text\"].apply(lambda x: \" \".join([stemmer.stem(word.strip()) for word in x.split() if word not in eng_stop_words if len(word) >= 3]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>text</th>\n",
       "      <th>class</th>\n",
       "      <th>formatted_text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>21st Century Wire says Ben Stein, reputable pr...</td>\n",
       "      <td>0</td>\n",
       "      <td>centuri wire say ben stein reput professor pep...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>WASHINGTON (Reuters) - U.S. President Donald T...</td>\n",
       "      <td>1</td>\n",
       "      <td>washington reuter presid donald trump remov ch...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>(Reuters) - Puerto Rico Governor Ricardo Rosse...</td>\n",
       "      <td>1</td>\n",
       "      <td>reuter puerto rico governor ricardo rossello s...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>On Monday, Donald Trump once again embarrassed...</td>\n",
       "      <td>0</td>\n",
       "      <td>monday donald trump embarrass countri accident...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>GLASGOW, Scotland (Reuters) - Most U.S. presid...</td>\n",
       "      <td>1</td>\n",
       "      <td>glasgow scotland reuter presidenti candid abro...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text  class  \\\n",
       "0  21st Century Wire says Ben Stein, reputable pr...      0   \n",
       "1  WASHINGTON (Reuters) - U.S. President Donald T...      1   \n",
       "2  (Reuters) - Puerto Rico Governor Ricardo Rosse...      1   \n",
       "3  On Monday, Donald Trump once again embarrassed...      0   \n",
       "4  GLASGOW, Scotland (Reuters) - Most U.S. presid...      1   \n",
       "\n",
       "                                      formatted_text  \n",
       "0  centuri wire say ben stein reput professor pep...  \n",
       "1  washington reuter presid donald trump remov ch...  \n",
       "2  reuter puerto rico governor ricardo rossello s...  \n",
       "3  monday donald trump embarrass countri accident...  \n",
       "4  glasgow scotland reuter presidenti candid abro...  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Saving cleaned and stemmed version of data\n",
    "# data.to_csv(\"Dataset/clean_data.csv\", index=False)\n",
    "# Reading in stemmed version\n",
    "data = pd.read_csv(\"Dataset/clean_data.csv\")\n",
    "data.drop(\"Unnamed: 0\", axis=1, inplace=True)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 44898 entries, 0 to 44897\n",
      "Data columns (total 3 columns):\n",
      " #   Column          Non-Null Count  Dtype \n",
      "---  ------          --------------  ----- \n",
      " 0   text            44898 non-null  object\n",
      " 1   class           44898 non-null  int64 \n",
      " 2   formatted_text  44266 non-null  object\n",
      "dtypes: int64(1), object(2)\n",
      "memory usage: 1.0+ MB\n"
     ]
    }
   ],
   "source": [
    "data.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "text                0\n",
       "class               0\n",
       "formatted_text    632\n",
       "dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.dropna(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "text              0\n",
       "class             0\n",
       "formatted_text    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Combining every Real news and Fake news into two distinct one to calculate count vectors of top words in real and fake news.\n",
    "\n",
    "# All reviews grouped by class\n",
    "all_formatted_text = data.groupby(\"class\")[\"formatted_text\"]\n",
    "\n",
    "# Extraction of distinct groups (Series)\n",
    "fake_news_series = all_formatted_text.get_group(0)\n",
    "real_news_series = all_formatted_text.get_group(1)\n",
    "\n",
    "# Creating lists containing all news\n",
    "fake_news_list = [news for news in fake_news_series]\n",
    "real_news_list = [news for news in real_news_series]\n",
    "all_news_list = fake_news_list + real_news_list\n",
    "\n",
    "# Creating string containing all words in each news type\n",
    "fake_news_string = \" \".join(fake_news_list)\n",
    "real_news_string = \" \".join(real_news_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "news_dictionary = {\n",
    "    \"fake\": fake_news_string,\n",
    "    \"real\": real_news_string\n",
    "}\n",
    "\n",
    "# Dataframe of all strings\n",
    "news_class_frame = pd.DataFrame(news_dictionary, index=[\"news_text\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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>news_text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>fake</th>\n",
       "      <td>centuri wire say ben stein reput professor pep...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>real</th>\n",
       "      <td>washington reuter presid donald trump remov ch...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                              news_text\n",
       "fake  centuri wire say ben stein reput professor pep...\n",
       "real  washington reuter presid donald trump remov ch..."
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "news_class_frame = news_class_frame.head().T\n",
    "news_class_frame.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "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>fake</th>\n",
       "      <th>real</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>trump</th>\n",
       "      <td>0.436805</td>\n",
       "      <td>0.286334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>said</th>\n",
       "      <td>0.177470</td>\n",
       "      <td>0.517953</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>presid</th>\n",
       "      <td>0.167065</td>\n",
       "      <td>0.157512</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>peopl</th>\n",
       "      <td>0.148863</td>\n",
       "      <td>0.080272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>one</th>\n",
       "      <td>0.139865</td>\n",
       "      <td>0.071905</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            fake      real\n",
       "trump   0.436805  0.286334\n",
       "said    0.177470  0.517953\n",
       "presid  0.167065  0.157512\n",
       "peopl   0.148863  0.080272\n",
       "one     0.139865  0.071905"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Calculating text frequencies\n",
    "transformer = TfidfVectorizer(ngram_range=(1,2))\n",
    "news_transformer = transformer.fit_transform(news_class_frame[\"news_text\"])\n",
    "\n",
    "\n",
    "# making a dataframe that contains all feature names and frequencies in order to determine new possible stop words.\n",
    "document_term_freq_df = pd.DataFrame(news_transformer.toarray(), columns= transformer.get_feature_names_out(), index=[\"fake\", \"real\"])\n",
    "document_term_freq_df = document_term_freq_df.T.sort_values(by= [\"fake\", \"real\"], ascending=False)\n",
    "document_term_freq_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Saving model as pickle file.\n",
    "pickle.dump(transformer, open(\"news-transformer.pkl\", \"wb\"))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> Based on the information above, the word **\"trump\"** is used a lot in both real and fake news however, not nearly as often therefore, it will not be placed as a stopword and other words are also not used nearly as often in both real and fake news."
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## List of data pre-processing/feature engineering steps taken.\n",
    "- Turning texts into lower case.\n",
    "- Removal of numbers.\n",
    "- Removal of symbols, special characters and white spaces.\n",
    "- Removal of stopwords (english).\n",
    "- stemming of words.\n",
    "- Text transformation using TF-IDF Vectorizer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Splitting data into train and test sets\n",
    "x = data[\"formatted_text\"]\n",
    "x = transformer.transform(x)\n",
    "y = data[\"class\"]\n",
    "\n",
    "x_train, x_test, y_train, y_test = train_test_split(x, y, test_size= 0.2, random_state= 42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;background-color: white;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>DecisionTreeClassifier()</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-2\" type=\"checkbox\" checked><label for=\"sk-estimator-id-2\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">DecisionTreeClassifier</label><div class=\"sk-toggleable__content\"><pre>DecisionTreeClassifier()</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "DecisionTreeClassifier()"
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Model training \n",
    "\n",
    "model = DecisionTreeClassifier()\n",
    "model.fit(x_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Saving model as pickle file\n",
    "pickle.dump(model, open(\"pred-model.pkl\", \"wb\"))"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The accuracy on Test data is 0.9994%\n",
      "The accuracy on Training data is 0.9995%\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "# Predicting test dataset\n",
    "test_predictions = model.predict(x_test)\n",
    "test_score = accuracy_score(y_test, test_predictions)\n",
    "\n",
    "train_predictions = model.predict(x_train)\n",
    "train_score = accuracy_score(y_train, train_predictions)\n",
    "\n",
    "# Printing results\n",
    "print(f\"The accuracy on Test data is {round(test_score, 4)}%\")\n",
    "print(f\"The accuracy on Training data is {round(train_score, 4)}%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Confusion matric plotting\n",
    "confusion_mat = confusion_matrix(y_test, test_predictions)\n",
    "sns.heatmap(confusion_mat, annot=True, fmt=\"d\", cbar=False)\n",
    "plt.title(\"Heatmap of news predictions\", fontsize=15)\n",
    "plt.show()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Final findings/summary\n",
    "\n",
    "- The model performs well on both training and test data with no huge difference therefore, The model is good."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0], dtype=int64)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pickle\n",
    "\n",
    "model = pickle.load(open(\"news-transformer.pkl\", \"rb\"))\n",
    "model_2 = pickle.load(open(\"pred-model.pkl\", \"rb\"))\n",
    "text = [\"This system is a conjunction of AI and NLP (Natural Language prediction) for fake news prediction. It employs the use of machine learning model trained with labeled data to classify news in these two broad categories: Real and Fake news.\"]\n",
    "\n",
    "res = model.transform(text)\n",
    "model_2.predict(res)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "fake-news-env",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.3"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "05f1a417725a930accbdf1a81f79fb64ecc94fb716920abfcaa49d2ff9c5d153"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}