12/31/23, 4:11 PM Fake News Classification.

ipynb - Colaboratory

keyboard_arrow_down Required Libraries

import pandas as pd
import numpy as np
import re
import nltk
from nltk.corpus import stopwords
from nltk.stem import PorterStemmer, WordNetLemmatizer
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics import accuracy_score, confusion_matrix, classification_report

keyboard_arrow_down 1. Data Gathering

df = pd.read_csv("/content/drive/MyDrive/Fake news detection/News_dataset.csv")
df.head()

id title author text label

House Dem Aide: We

House Dem Aide: We Didn’t
0 0 Darrell Lucus Didn’t Even See Comey’s 1
Even See Comey’s Let...
Let...

FLYNN: Hillary Clinton, Big Ever get the feeling your

1 1 Daniel J. Flynn 0
Woman on Campus - ... life circles the rou...

Why the Truth Might Get You Why the Truth Might Get
2 2 Consortiumnews.com 1
Fired You Fired October 29, ...

keyboard_arrow_down 2. Data Analysis

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 20800 entries, 0 to 20799
Data columns (total 5 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 id 20800 non-null int64
1 title 20242 non-null object
2 author 18843 non-null object
3 text 20761 non-null object
4 label 20800 non-null int64
dtypes: int64(2), object(3)
memory usage: 812.6+ KB

df['label'].value_counts()

1 10413
0 10387
Name: label, dtype: int64

df.shape

(20800, 5)

df.isna().sum()

id 0
title 558
author 1957
text 39
label 0
dtype: int64

df = df.dropna() #Handled Missing values by droping those rows

df.isna().sum()

id 0
title 0

https://colab.research.google.com/drive/196NFXnYInNj9bFCE3lyCRGStXYB71vPs?authuser=2#printMode=true 1/6
12/31/23, 4:11 PM Fake News Classification.ipynb - Colaboratory
author 0
text 0
label 0
dtype: int64

df.shape

(18285, 5)

df.reset_index(inplace=True)
df.head()

index id title author text label

House Dem Aide: We House Dem Aide: We

0 0 0 Didn’t Even See Comey’s Darrell Lucus Didn’t Even See 1
Let... Comey’s Let...

FLYNN: Hillary Clinton, Ever get the feeling

1 1 1 Big Woman on Campus - Daniel J. Flynn your life circles the 0
... rou...

Why the Truth Might

Why the Truth Might Get
2 2 2 Consortiumnews com Get You Fired October 1

df['title'][0]

'House Dem Aide: We Didn’t Even See Comey’s Letter Until Jason Chaffetz Tweeted It'

df = df.drop(['id','text','author'],axis = 1)
df.head()

index title label

0 0 House Dem Aide: We Didn’t Even See Comey’s Let... 1

1 1 FLYNN: Hillary Clinton, Big Woman on Campus - ... 0

2 2 Why the Truth Might Get You Fired 1

3 3 15 Civilians Killed In Single US Airstrike Hav... 1

4 4 Iranian woman jailed for fictional unpublished... 1

keyboard_arrow_down 3. Data Preprocessing

keyboard_arrow_down 1.Tokenization
sample_data = 'The quick brown fox jumps over the lazy dog'
sample_data = sample_data.split()
sample_data

['The', 'quick', 'brown', 'fox', 'jumps', 'over', 'the', 'lazy', 'dog']

keyboard_arrow_down 2. Make Lowercase

sample_data = [data.lower() for data in sample_data]
sample_data

['the', 'quick', 'brown', 'fox', 'jumps', 'over', 'the', 'lazy', 'dog']

keyboard_arrow_down 3. Remove Stopwords

nltk.download('stopwords')
stopwords = stopwords.words('english')
print(stopwords[0:10])
print(len(stopwords))

['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', "you're"]
179
[nltk_data] Downloading package stopwords to /root/nltk_data...
[nltk_data] Unzipping corpora/stopwords.zip.

https://colab.research.google.com/drive/196NFXnYInNj9bFCE3lyCRGStXYB71vPs?authuser=2#printMode=true 2/6
12/31/23, 4:11 PM Fake News Classification.ipynb - Colaboratory
sample_data = [data for data in sample_data if data not in stopwords]
print(sample_data)
len(sample_data)

['quick', 'brown', 'fox', 'jumps', 'lazy', 'dog']

6

keyboard_arrow_down 4. Stemming
ps = PorterStemmer()
sample_data_stemming = [ps.stem(data) for data in sample_data]
print(sample_data_stemming)

['quick', 'brown', 'fox', 'jump', 'lazi', 'dog']

keyboard_arrow_down 5. Lemmatization
nltk.download('wordnet')
lm = WordNetLemmatizer()
sample_data_lemma = [lm.lemmatize(data) for data in sample_data]
print(sample_data_lemma)

[nltk_data] Downloading package wordnet to /root/nltk_data...

['quick', 'brown', 'fox', 'jump', 'lazy', 'dog']

lm = WordNetLemmatizer()
corpus = []
for i in range (len(df)):
review = re.sub('^a-zA-Z0-9',' ', df['title'][i])
review = review.lower()
review = review.split()
review = [lm.lemmatize(x) for x in review if x not in stopwords]
review = " ".join(review)
corpus.append(review)

len(corpus)

18285

df['title'][0]

'House Dem Aide: We Didn’t Even See Comey’s Letter Until Jason Chaffetz Tweeted It'

corpus[0]

'house dem aide: didn’t even see comey’s letter jason chaffetz tweeted'

keyboard_arrow_down 4. Vectorization (Convert Text data into the Vector)

tf = TfidfVectorizer()
x = tf.fit_transform(corpus).toarray()
x

array([[0., 0., 0., ..., 0., 0., 0.],

[0., 0., 0., ..., 0., 0., 0.],
[0., 0., 0., ..., 0., 0., 0.],
...,
[0., 0., 0., ..., 0., 0., 0.],
[0., 0., 0., ..., 0., 0., 0.],
[0., 0., 0., ..., 0., 0., 0.]])

y = df['label']
y.head()

0 1
1 0
2 1
3 1
4 1
Name: label, dtype: int64

https://colab.research.google.com/drive/196NFXnYInNj9bFCE3lyCRGStXYB71vPs?authuser=2#printMode=true 3/6
12/31/23, 4:11 PM Fake News Classification.ipynb - Colaboratory

keyboard_arrow_down Data splitting into the train and test

x_train, x_test, y_train, y_test = train_test_split(x,y, test_size = 0.3, random_state = 10, stratify = y )

len(x_train),len(y_train)

(12799, 12799)

len(x_test), len(y_test)

(5486, 5486)

keyboard_arrow_down 5. Model Building

rf = RandomForestClassifier()
rf.fit(x_train, y_train)

▾ RandomForestClassifier
RandomForestClassifier()

keyboard_arrow_down 6. Model Evaluation

y_pred = rf.predict(x_test)
accuracy_score_ = accuracy_score(y_test,y_pred)
accuracy_score_

0.9374772147283995

class Evaluation:

def __init__(self,model,x_train,x_test,y_train,y_test):
self.model = model
self.x_train = x_train
self.x_test = x_test
self.y_train = y_train
self.y_test = y_test

def train_evaluation(self):
y_pred_train = self.model.predict(self.x_train)

acc_scr_train = accuracy_score(self.y_train,y_pred_train)
print("Accuracy Score On Training Data Set :",acc_scr_train)
print()

con_mat_train = confusion_matrix(self.y_train,y_pred_train)
print("Confusion Matrix On Training Data Set :\n",con_mat_train)
print()

class_rep_train = classification_report(self.y_train,y_pred_train)
print("Classification Report On Training Data Set :\n",class_rep_train)

def test_evaluation(self):
y_pred_test = self.model.predict(self.x_test)

acc_scr_test = accuracy_score(self.y_test,y_pred_test)
print("Accuracy Score On Testing Data Set :",acc_scr_test)
print()

con_mat_test = confusion_matrix(self.y_test,y_pred_test)
print("Confusion Matrix On Testing Data Set :\n",con_mat_test)
print()

class_rep_test = classification_report(self.y_test,y_pred_test)
print("Classification Report On Testing Data Set :\n",class_rep_test)

https://colab.research.google.com/drive/196NFXnYInNj9bFCE3lyCRGStXYB71vPs?authuser=2#printMode=true 4/6
12/31/23, 4:11 PM Fake News Classification.ipynb - Colaboratory
#Checking the accuracy on training dataset

Evaluation(rf,x_train, x_test, y_train, y_test).train_evaluation()

Accuracy Score On Training Data Set : 1.0

Confusion Matrix On Training Data Set :

[[7252 0]
[ 0 5547]]

Classification Report On Training Data Set :

precision recall f1-score support

0 1.00 1.00 1.00 7252

1 1.00 1.00 1.00 5547

accuracy 1.00 12799

macro avg 1.00 1.00 1.00 12799
weighted avg 1.00 1.00 1.00 12799

#Checking the accuracy on testing dataset

Evaluation(rf,x_train, x_test, y_train, y_test).test_evaluation()

Accuracy Score On Testing Data Set : 0.9374772147283995

Confusion Matrix On Testing Data Set :

[[2825 284]
[ 59 2318]]

Classification Report On Testing Data Set :

precision recall f1-score support

0 0.98 0.91 0.94 3109

1 0.89 0.98 0.93 2377

accuracy 0.94 5486

macro avg 0.94 0.94 0.94 5486
weighted avg 0.94 0.94 0.94 5486

keyboard_arrow_down Prediction Pipeline

class Preprocessing:

def __init__(self,data):
self.data = data

def text_preprocessing_user(self):
lm = WordNetLemmatizer()
pred_data = [self.data]
preprocess_data = []
for data in pred_data:
review = re.sub('^a-zA-Z0-9',' ', data)
review = review.lower()
review = review.split()
review = [lm.lemmatize(x) for x in review if x not in stopwords]
review = " ".join(review)
preprocess_data.append(review)
return preprocess_data

df['title'][1]

'FLYNN: Hillary Clinton, Big Woman on Campus - Breitbart'

data = 'FLYNN: Hillary Clinton, Big Woman on Campus - Breitbart'

Preprocessing(data).text_preprocessing_user()

['flynn: hillary clinton, big woman campus - breitbart']

class Prediction:

def __init__(self,pred_data, model):

self.pred_data = pred_data
self.model = model

def prediction_model(self):
preprocess_data = Preprocessing(self.pred_data).text_preprocessing_user()
data = tf.transform(preprocess_data)

https://colab.research.google.com/drive/196NFXnYInNj9bFCE3lyCRGStXYB71vPs?authuser=2#printMode=true 5/6
12/31/23, 4:11 PM Fake News Classification.ipynb - Colaboratory
prediction = self.model.predict(data)

if prediction [0] == 0 :
return "The News Is Fake"

else:
return "The News Is Real"

data = 'FLYNN: Hillary Clinton, Big Woman on Campus - Breitbart'

Prediction(data,rf).prediction_model()

'The News Is Fake'

df['title'][3]

'15 Civilians Killed In Single US Airstrike Have Been Identified'

user_data = '15 Civilians Killed In Single US Airstrike Have Been Identified'

Prediction(user_data,rf).prediction_model()

'The News Is Real'

https://colab.research.google.com/drive/196NFXnYInNj9bFCE3lyCRGStXYB71vPs?authuser=2#printMode=true 6/6