Text_Representation_Using_NGrams.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4fa3e978-ccc4-4fa0-804a-b05f1cd1dcff",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'chem': 1, 'the': 6, 'charming': 0, 'is': 3, 'looking': 5, 'for': 2, 'job': 4}"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "\n",
    "v = CountVectorizer() # Convert a collection of text documents to a matrix of token counts.\n",
    "\n",
    "v.fit([\"Chem the charming is looking for a job\"])\n",
    "# the fit method \"learn(s) a vocabulary dictionary of all tokens in the raw documents\n",
    "#  it creates a dictionary of tokens (by default the tokens are words\n",
    "v.vocabulary_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c53c14d2-5e8f-4914-9902-4d86fbfee46a",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'chem': 2,\n",
       " 'the': 11,\n",
       " 'charming': 0,\n",
       " 'is': 6,\n",
       " 'looking': 9,\n",
       " 'for': 4,\n",
       " 'job': 8,\n",
       " 'chem the': 3,\n",
       " 'the charming': 12,\n",
       " 'charming is': 1,\n",
       " 'is looking': 7,\n",
       " 'looking for': 10,\n",
       " 'for job': 5}"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "V = CountVectorizer(ngram_range=(1,2)) # i.e. ngram_range creates a bag of words of 1 & 2 as given\n",
    "V.fit([\"Chem the charming is looking for a job\"])\n",
    "V.vocabulary_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "ac473167-e6bd-4375-b5c3-a19a219226f4",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'chem': 3,\n",
       " 'the': 15,\n",
       " 'charming': 0,\n",
       " 'is': 8,\n",
       " 'looking': 12,\n",
       " 'for': 6,\n",
       " 'job': 11,\n",
       " 'chem the': 4,\n",
       " 'the charming': 16,\n",
       " 'charming is': 1,\n",
       " 'is looking': 9,\n",
       " 'looking for': 13,\n",
       " 'for job': 7,\n",
       " 'chem the charming': 5,\n",
       " 'the charming is': 17,\n",
       " 'charming is looking': 2,\n",
       " 'is looking for': 10,\n",
       " 'looking for job': 14}"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Bag of words in the range of 1 and 3 tokens\n",
    "V = CountVectorizer(ngram_range=(1,3))\n",
    "V.fit([\"Chem the charming is looking for a job\"])\n",
    "V.vocabulary_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "25aa6975-e6cb-40ed-a654-6f203e983c5e",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Corpus of words\n",
    "corpus = [\n",
    "    \"Chem is charming guy\",\n",
    "    \"Chem likes to eat stuff from bakery\",\n",
    "    \"Chem also likes wine\"\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "dcfffa82-3d95-4544-9235-2e481c76d30e",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import spacy\n",
    "\n",
    "nlp = spacy.load(\"en_core_web_sm\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d62b6489-83ef-49fe-a2bc-bc9e4cf4eba6",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Lemmatize and remove the stopwords\n",
    "def Preprocess(text):\n",
    "    doc = nlp(text)\n",
    "    filtered_tokens = []\n",
    "    for token in doc:\n",
    "        if token.is_punct or token.is_stop:\n",
    "            continue\n",
    "        filtered_tokens.append(token.lemma_)\n",
    "    \n",
    "    return \" \".join(filtered_tokens)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "1e828537-eb25-4d92-be5a-ae7267677deb",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Chem charming guy'"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Preprocess(\"Chem is a charming guy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "aba68b09-a38b-4d9a-893f-7cb02174c34c",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Chem charming guy', 'chem like eat stuff bakery', 'Chem like wine']"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corpus_processed = [Preprocess(token) for token in corpus]\n",
    "corpus_processed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "99a919db-49d0-4590-8626-c30899b35039",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'chem': 3,\n",
       " 'charming': 1,\n",
       " 'guy': 12,\n",
       " 'chem charming': 4,\n",
       " 'charming guy': 2,\n",
       " 'chem charming guy': 5,\n",
       " 'like': 13,\n",
       " 'eat': 9,\n",
       " 'stuff': 17,\n",
       " 'bakery': 0,\n",
       " 'chem like': 6,\n",
       " 'like eat': 14,\n",
       " 'eat stuff': 10,\n",
       " 'stuff bakery': 18,\n",
       " 'chem like eat': 7,\n",
       " 'like eat stuff': 15,\n",
       " 'eat stuff bakery': 11,\n",
       " 'wine': 19,\n",
       " 'like wine': 16,\n",
       " 'chem like wine': 8}"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "V.fit(corpus_processed)\n",
    "V.vocabulary_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "d40131c9-6ad1-4490-8d3e-04b1dd2a943f",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0]],\n",
       "      dtype=int64)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Generate a bag of n gram vector for few sample documents\n",
    "V.transform([\"Chem is a charming guy\"]).toarray()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "8c3320be-4592-4265-af34-a19e54c3993d",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(209527, 6)\n"
     ]
    },
    {
     "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>link</th>\n",
       "      <th>headline</th>\n",
       "      <th>category</th>\n",
       "      <th>short_description</th>\n",
       "      <th>authors</th>\n",
       "      <th>date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>https://www.huffpost.com/entry/covid-boosters-...</td>\n",
       "      <td>Over 4 Million Americans Roll Up Sleeves For O...</td>\n",
       "      <td>U.S. NEWS</td>\n",
       "      <td>Health experts said it is too early to predict...</td>\n",
       "      <td>Carla K. Johnson, AP</td>\n",
       "      <td>2022-09-23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>https://www.huffpost.com/entry/american-airlin...</td>\n",
       "      <td>American Airlines Flyer Charged, Banned For Li...</td>\n",
       "      <td>U.S. NEWS</td>\n",
       "      <td>He was subdued by passengers and crew when he ...</td>\n",
       "      <td>Mary Papenfuss</td>\n",
       "      <td>2022-09-23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>https://www.huffpost.com/entry/funniest-tweets...</td>\n",
       "      <td>23 Of The Funniest Tweets About Cats And Dogs ...</td>\n",
       "      <td>COMEDY</td>\n",
       "      <td>\"Until you have a dog you don't understand wha...</td>\n",
       "      <td>Elyse Wanshel</td>\n",
       "      <td>2022-09-23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>https://www.huffpost.com/entry/funniest-parent...</td>\n",
       "      <td>The Funniest Tweets From Parents This Week (Se...</td>\n",
       "      <td>PARENTING</td>\n",
       "      <td>\"Accidentally put grown-up toothpaste on my to...</td>\n",
       "      <td>Caroline Bologna</td>\n",
       "      <td>2022-09-23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>https://www.huffpost.com/entry/amy-cooper-lose...</td>\n",
       "      <td>Woman Who Called Cops On Black Bird-Watcher Lo...</td>\n",
       "      <td>U.S. NEWS</td>\n",
       "      <td>Amy Cooper accused investment firm Franklin Te...</td>\n",
       "      <td>Nina Golgowski</td>\n",
       "      <td>2022-09-22</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                link  \\\n",
       "0  https://www.huffpost.com/entry/covid-boosters-...   \n",
       "1  https://www.huffpost.com/entry/american-airlin...   \n",
       "2  https://www.huffpost.com/entry/funniest-tweets...   \n",
       "3  https://www.huffpost.com/entry/funniest-parent...   \n",
       "4  https://www.huffpost.com/entry/amy-cooper-lose...   \n",
       "\n",
       "                                            headline   category  \\\n",
       "0  Over 4 Million Americans Roll Up Sleeves For O...  U.S. NEWS   \n",
       "1  American Airlines Flyer Charged, Banned For Li...  U.S. NEWS   \n",
       "2  23 Of The Funniest Tweets About Cats And Dogs ...     COMEDY   \n",
       "3  The Funniest Tweets From Parents This Week (Se...  PARENTING   \n",
       "4  Woman Who Called Cops On Black Bird-Watcher Lo...  U.S. NEWS   \n",
       "\n",
       "                                   short_description               authors  \\\n",
       "0  Health experts said it is too early to predict...  Carla K. Johnson, AP   \n",
       "1  He was subdued by passengers and crew when he ...        Mary Papenfuss   \n",
       "2  \"Until you have a dog you don't understand wha...         Elyse Wanshel   \n",
       "3  \"Accidentally put grown-up toothpaste on my to...      Caroline Bologna   \n",
       "4  Amy Cooper accused investment firm Franklin Te...        Nina Golgowski   \n",
       "\n",
       "        date  \n",
       "0 2022-09-23  \n",
       "1 2022-09-23  \n",
       "2 2022-09-23  \n",
       "3 2022-09-23  \n",
       "4 2022-09-22  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "df = pd.read_json('News_Dataset.json',lines=True)\n",
    "print(df.shape)\n",
    "\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "08c2af76-440c-43fc-a091-1b020de6d7b2",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "POLITICS          35602\n",
       "WELLNESS          17945\n",
       "ENTERTAINMENT     17362\n",
       "TRAVEL             9900\n",
       "STYLE & BEAUTY     9814\n",
       "PARENTING          8791\n",
       "HEALTHY LIVING     6694\n",
       "QUEER VOICES       6347\n",
       "FOOD & DRINK       6340\n",
       "BUSINESS           5992\n",
       "COMEDY             5400\n",
       "SPORTS             5077\n",
       "BLACK VOICES       4583\n",
       "HOME & LIVING      4320\n",
       "PARENTS            3955\n",
       "THE WORLDPOST      3664\n",
       "WEDDINGS           3653\n",
       "WOMEN              3572\n",
       "CRIME              3562\n",
       "IMPACT             3484\n",
       "DIVORCE            3426\n",
       "WORLD NEWS         3299\n",
       "MEDIA              2944\n",
       "WEIRD NEWS         2777\n",
       "GREEN              2622\n",
       "WORLDPOST          2579\n",
       "RELIGION           2577\n",
       "STYLE              2254\n",
       "SCIENCE            2206\n",
       "TECH               2104\n",
       "TASTE              2096\n",
       "MONEY              1756\n",
       "ARTS               1509\n",
       "ENVIRONMENT        1444\n",
       "FIFTY              1401\n",
       "GOOD NEWS          1398\n",
       "U.S. NEWS          1377\n",
       "ARTS & CULTURE     1339\n",
       "COLLEGE            1144\n",
       "LATINO VOICES      1130\n",
       "CULTURE & ARTS     1074\n",
       "EDUCATION          1014\n",
       "Name: category, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.category.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "64d780b0-afa0-4bfa-9564-729f2cd8ea48",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Handling class imbalance\n",
    "\n",
    "min_samples = 8791\n",
    "\n",
    "df_Parenting = df[df.category=='PARENTING'].sample(min_samples,random_state=2022)\n",
    "df_Travel = df[df.category=='TRAVEL'].sample(min_samples,random_state=2022)\n",
    "df_Entertainment = df[df.category=='ENTERTAINMENT'].sample(min_samples,random_state=2022)\n",
    "df_Wellness = df[df.category=='WELLNESS'].sample(min_samples,random_state=2022)\n",
    "df_Politics = df[df.category=='POLITICS'].sample(min_samples,random_state=2022)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "03fa8cc4-4826-4a4b-ab32-d9a33d22a270",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PARENTING        8791\n",
       "TRAVEL           8791\n",
       "ENTERTAINMENT    8791\n",
       "WELLNESS         8791\n",
       "POLITICS         8791\n",
       "Name: category, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_balanced = pd.concat([df_Parenting,df_Travel,df_Entertainment,df_Wellness,df_Politics],axis=0)\n",
    "df_balanced.category.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "1d56ca27-c0ce-46e0-b1c2-27673f04ea6d",
   "metadata": {
    "tags": []
   },
   "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>link</th>\n",
       "      <th>headline</th>\n",
       "      <th>category</th>\n",
       "      <th>short_description</th>\n",
       "      <th>authors</th>\n",
       "      <th>date</th>\n",
       "      <th>Category_num</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>154193</th>\n",
       "      <td>https://www.huffingtonpost.comhttp://ideas.tim...</td>\n",
       "      <td>Bringing Up a 9/11 Baby</td>\n",
       "      <td>PARENTING</td>\n",
       "      <td>One year ago today, at 11:45 pm, my wife gave ...</td>\n",
       "      <td></td>\n",
       "      <td>2013-09-11</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>201170</th>\n",
       "      <td>https://www.huffingtonpost.com/entry/ethan-pia...</td>\n",
       "      <td>Ethan Walmark, 6-Year-Old With Autism, Plays '...</td>\n",
       "      <td>PARENTING</td>\n",
       "      <td>Ethan Walmark, 6, was diagnosed with autism fo...</td>\n",
       "      <td>Jessica Samakow</td>\n",
       "      <td>2012-04-27</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>140458</th>\n",
       "      <td>https://www.huffingtonpost.com/entry/how-to-lo...</td>\n",
       "      <td>How to Lose an Argument With a 5-Year-Old</td>\n",
       "      <td>PARENTING</td>\n",
       "      <td>Listen to me closely, mister. From this point ...</td>\n",
       "      <td>Wendi Aarons, Contributor\\nHumor writer, blogg...</td>\n",
       "      <td>2014-02-03</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                     link  \\\n",
       "154193  https://www.huffingtonpost.comhttp://ideas.tim...   \n",
       "201170  https://www.huffingtonpost.com/entry/ethan-pia...   \n",
       "140458  https://www.huffingtonpost.com/entry/how-to-lo...   \n",
       "\n",
       "                                                 headline   category  \\\n",
       "154193                            Bringing Up a 9/11 Baby  PARENTING   \n",
       "201170  Ethan Walmark, 6-Year-Old With Autism, Plays '...  PARENTING   \n",
       "140458          How to Lose an Argument With a 5-Year-Old  PARENTING   \n",
       "\n",
       "                                        short_description  \\\n",
       "154193  One year ago today, at 11:45 pm, my wife gave ...   \n",
       "201170  Ethan Walmark, 6, was diagnosed with autism fo...   \n",
       "140458  Listen to me closely, mister. From this point ...   \n",
       "\n",
       "                                                  authors       date  \\\n",
       "154193                                                    2013-09-11   \n",
       "201170                                    Jessica Samakow 2012-04-27   \n",
       "140458  Wendi Aarons, Contributor\\nHumor writer, blogg... 2014-02-03   \n",
       "\n",
       "        Category_num  \n",
       "154193             0  \n",
       "201170             0  \n",
       "140458             0  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_balanced['Category_num'] = df_balanced['category'].map({\n",
    "    'PARENTING':0,\n",
    "    'TRAVEL':1,\n",
    "    'ENTERTAINMENT':2,\n",
    "    'WELLNESS':3,\n",
    "    'POLITICS':4,})\n",
    "df_balanced.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "d09f6a63-9cde-4af3-90c6-1e0b46dcf725",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    df_balanced.short_description, \n",
    "    df_balanced.Category_num, \n",
    "    test_size=0.2, # 20% samples will go to test dataset\n",
    "    random_state=2022,\n",
    "    stratify=df_balanced.Category_num\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "62734993-0046-48b6-94d0-911b020af4b2",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(35164,)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "128846                                                     \n",
       "148225    I make my big declarations of gratitude at Tha...\n",
       "198654    Lightly lined, seamless bandeau, structured ba...\n",
       "43637     After many delays, the Senate can now proceed ...\n",
       "162076    This past Memorial Day weekend, we became “tha...\n",
       "Name: short_description, dtype: object"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(X_train.shape)\n",
    "X_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "fdda888b-786c-4825-bc39-d5e2d7c6d994",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2    7033\n",
       "0    7033\n",
       "3    7033\n",
       "1    7033\n",
       "4    7032\n",
       "Name: Category_num, dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "c7f01ebc-dd37-4c0d-8d07-1da0286f8c90",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.50      0.78      0.61      1758\n",
      "           1       0.80      0.74      0.77      1758\n",
      "           2       0.80      0.45      0.58      1758\n",
      "           3       0.68      0.79      0.73      1758\n",
      "           4       0.84      0.67      0.74      1759\n",
      "\n",
      "    accuracy                           0.69      8791\n",
      "   macro avg       0.72      0.69      0.69      8791\n",
      "weighted avg       0.72      0.69      0.69      8791\n",
      "\n"
     ]
    }
   ],
   "source": [
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.metrics import classification_report\n",
    "\n",
    "#1. create a pipeline object\n",
    "clf = Pipeline([\n",
    "     ('vectorizer_bow', CountVectorizer(ngram_range = (1, 1))),        #using the ngram_range parameter \n",
    "     ('Multi NB', MultinomialNB())         \n",
    "])\n",
    "\n",
    "#2. fit with X_train and y_train\n",
    "clf.fit(X_train, y_train)\n",
    "\n",
    "\n",
    "#3. get the predictions for X_test and store it in y_pred\n",
    "y_pred = clf.predict(X_test)\n",
    "\n",
    "\n",
    "#4. print the classfication report\n",
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "6d8083b0-a5a5-418e-896d-46732432bfb4",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "25922     Hope you had a nice long weekend, this is the ...\n",
       "103888    Flight attendants take on a crucial role after...\n",
       "194877    Tom Cruise's personal life may be in shambles,...\n",
       "137804    With all the DOs and DON’Ts of dieting, losing...\n",
       "196331    We tend to think mistakes get in the way of pr...\n",
       "Name: short_description, dtype: object"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_test[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "721abe63-06ea-4e06-867c-b5ce01adc1ec",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([4, 1, 4, 3, 3], dtype=int64)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_pred[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "192b677b-f49e-4d0c-b6f0-af6153127ae9",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "25922     4\n",
       "103888    1\n",
       "194877    2\n",
       "137804    3\n",
       "196331    3\n",
       "Name: Category_num, dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "2d23c001-6d18-4384-b67f-e6dcad95120a",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.47      0.81      0.60      1758\n",
      "           1       0.76      0.72      0.74      1758\n",
      "           2       0.90      0.31      0.47      1758\n",
      "           3       0.63      0.83      0.72      1758\n",
      "           4       0.87      0.59      0.70      1759\n",
      "\n",
      "    accuracy                           0.65      8791\n",
      "   macro avg       0.73      0.65      0.64      8791\n",
      "weighted avg       0.73      0.65      0.64      8791\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#1. create a pipeline object\n",
    "clf = Pipeline([\n",
    "    ('vectorizer_1_2_gram', CountVectorizer(ngram_range = (1, 2))),        #using the ngram_range parameter \n",
    "     ('Multi NB', MultinomialNB())         \n",
    "])\n",
    "\n",
    "#2. fit with X_train and y_train\n",
    "clf.fit(X_train, y_train)\n",
    "\n",
    "\n",
    "#3. get the predictions for X_test and store it in y_pred\n",
    "y_pred = clf.predict(X_test)\n",
    "\n",
    "\n",
    "#4. print the classfication report\n",
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "82989d59-7412-4b78-9644-eaa3388d7312",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.46      0.80      0.58      1758\n",
      "           1       0.74      0.71      0.72      1758\n",
      "           2       0.91      0.27      0.41      1758\n",
      "           3       0.60      0.82      0.69      1758\n",
      "           4       0.88      0.56      0.68      1759\n",
      "\n",
      "    accuracy                           0.63      8791\n",
      "   macro avg       0.72      0.63      0.62      8791\n",
      "weighted avg       0.72      0.63      0.62      8791\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#1. create a pipeline object\n",
    "clf = Pipeline([\n",
    "    ('vectorizer_1_3_grams', CountVectorizer(ngram_range = (1, 3))),        #using the ngram_range parameter \n",
    "     ('Multi NB', MultinomialNB())         \n",
    "])\n",
    "\n",
    "#2. fit with X_train and y_train\n",
    "clf.fit(X_train, y_train)\n",
    "\n",
    "\n",
    "#3. get the predictions for X_test and store it in y_pred\n",
    "y_pred = clf.predict(X_test)\n",
    "\n",
    "\n",
    "#4. print the classfication report\n",
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "a5b55e1f-aebd-4526-997f-cfabf2ee2f54",
   "metadata": {
    "tags": []
   },
   "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>link</th>\n",
       "      <th>headline</th>\n",
       "      <th>category</th>\n",
       "      <th>short_description</th>\n",
       "      <th>authors</th>\n",
       "      <th>date</th>\n",
       "      <th>Category_num</th>\n",
       "      <th>preprocessed_txt</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>154193</th>\n",
       "      <td>https://www.huffingtonpost.comhttp://ideas.tim...</td>\n",
       "      <td>Bringing Up a 9/11 Baby</td>\n",
       "      <td>PARENTING</td>\n",
       "      <td>One year ago today, at 11:45 pm, my wife gave ...</td>\n",
       "      <td></td>\n",
       "      <td>2013-09-11</td>\n",
       "      <td>0</td>\n",
       "      <td>year ago today 11:45 pm wife give birth son Ba...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>201170</th>\n",
       "      <td>https://www.huffingtonpost.com/entry/ethan-pia...</td>\n",
       "      <td>Ethan Walmark, 6-Year-Old With Autism, Plays '...</td>\n",
       "      <td>PARENTING</td>\n",
       "      <td>Ethan Walmark, 6, was diagnosed with autism fo...</td>\n",
       "      <td>Jessica Samakow</td>\n",
       "      <td>2012-04-27</td>\n",
       "      <td>0</td>\n",
       "      <td>Ethan Walmark 6 diagnose autism year ago feel ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>140458</th>\n",
       "      <td>https://www.huffingtonpost.com/entry/how-to-lo...</td>\n",
       "      <td>How to Lose an Argument With a 5-Year-Old</td>\n",
       "      <td>PARENTING</td>\n",
       "      <td>Listen to me closely, mister. From this point ...</td>\n",
       "      <td>Wendi Aarons, Contributor\\nHumor writer, blogg...</td>\n",
       "      <td>2014-02-03</td>\n",
       "      <td>0</td>\n",
       "      <td>listen closely mister point long argue want ac...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                     link  \\\n",
       "154193  https://www.huffingtonpost.comhttp://ideas.tim...   \n",
       "201170  https://www.huffingtonpost.com/entry/ethan-pia...   \n",
       "140458  https://www.huffingtonpost.com/entry/how-to-lo...   \n",
       "\n",
       "                                                 headline   category  \\\n",
       "154193                            Bringing Up a 9/11 Baby  PARENTING   \n",
       "201170  Ethan Walmark, 6-Year-Old With Autism, Plays '...  PARENTING   \n",
       "140458          How to Lose an Argument With a 5-Year-Old  PARENTING   \n",
       "\n",
       "                                        short_description  \\\n",
       "154193  One year ago today, at 11:45 pm, my wife gave ...   \n",
       "201170  Ethan Walmark, 6, was diagnosed with autism fo...   \n",
       "140458  Listen to me closely, mister. From this point ...   \n",
       "\n",
       "                                                  authors       date  \\\n",
       "154193                                                    2013-09-11   \n",
       "201170                                    Jessica Samakow 2012-04-27   \n",
       "140458  Wendi Aarons, Contributor\\nHumor writer, blogg... 2014-02-03   \n",
       "\n",
       "        Category_num                                   preprocessed_txt  \n",
       "154193             0  year ago today 11:45 pm wife give birth son Ba...  \n",
       "201170             0  Ethan Walmark 6 diagnose autism year ago feel ...  \n",
       "140458             0  listen closely mister point long argue want ac...  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_balanced['preprocessed_txt'] = df_balanced['short_description'].apply(Preprocess)\n",
    "df_balanced.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "1aef9bd4-0b70-4c0b-be82-b6c91a7ca737",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(35164,)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "128846                                                     \n",
       "148225    big declaration gratitude Thanksgiving dinner ...\n",
       "198654    lightly lined seamless bandeau structured band...\n",
       "43637     delay Senate proceed confirmation process fast...\n",
       "162076    past Memorial Day weekend family swear tv fami...\n",
       "Name: preprocessed_txt, dtype: object"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    df_balanced.preprocessed_txt, \n",
    "    df_balanced.Category_num, \n",
    "    test_size=0.2, # 20% samples will go to test dataset\n",
    "    random_state=2022,\n",
    "    stratify=df_balanced.Category_num\n",
    ")\n",
    "print(X_train.shape)\n",
    "X_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "4826715d-6347-443b-b8ac-f41e0a0e8a72",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2    7033\n",
       "0    7033\n",
       "3    7033\n",
       "1    7033\n",
       "4    7032\n",
       "Name: Category_num, dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_train.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "61ef2f24-3f17-4d64-8854-02b858fbbae8",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4    1759\n",
       "1    1758\n",
       "2    1758\n",
       "3    1758\n",
       "0    1758\n",
       "Name: Category_num, dtype: int64"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y_test.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "62458af8-28aa-4a51-9650-c9ec9e7f3917",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.50      0.80      0.62      1758\n",
      "           1       0.78      0.77      0.78      1758\n",
      "           2       0.89      0.37      0.52      1758\n",
      "           3       0.67      0.82      0.74      1758\n",
      "           4       0.87      0.67      0.76      1759\n",
      "\n",
      "    accuracy                           0.69      8791\n",
      "   macro avg       0.74      0.69      0.68      8791\n",
      "weighted avg       0.74      0.69      0.68      8791\n",
      "\n"
     ]
    }
   ],
   "source": [
    "#1. create a pipeline object\n",
    "clf = Pipeline([\n",
    "    ('vectorizer_bow', CountVectorizer(ngram_range = (1, 2))),        #using the ngram_range parameter \n",
    "    ('Multi NB', MultinomialNB())\n",
    "])\n",
    "\n",
    "#2. fit with X_train and y_train\n",
    "clf.fit(X_train, y_train)\n",
    "\n",
    "\n",
    "#3. get the predictions for X_test and store it in y_pred\n",
    "y_pred = clf.predict(X_test)\n",
    "\n",
    "\n",
    "#4. print the classfication report\n",
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "6ec8f5b6-97c5-4e88-9db8-4256cca1917e",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1406,   86,   16,  230,   20],\n",
       "       [ 226, 1362,   16,  134,   20],\n",
       "       [ 649,  158,  647,  184,  120],\n",
       "       [ 226,   61,   11, 1442,   18],\n",
       "       [ 302,   89,   39,  150, 1179]], dtype=int64)"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "cm = confusion_matrix(y_test, y_pred)\n",
    "cm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "29b519df-cd1f-40a8-baff-dd7ea66ebf2f",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(95.72222222222221, 0.5, 'Truth')"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x700 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot as plt\n",
    "import seaborn as sn\n",
    "plt.figure(figsize = (10,7))\n",
    "sn.heatmap(cm, annot=True, fmt='d')\n",
    "plt.xlabel('Prediction')\n",
    "plt.ylabel('Truth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b173b026-22b5-43cc-98d6-8bdbe122c249",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}