Fake News Detection Using Machine Learning in Python

By Akarsh Shekhar

In this tutorial program, we will learn about building fake news detector using machine learning with the language used is Python.  So here I am going to discuss what are the basic steps of this machine learning problem and how to approach it.

For fake news predictor, we are going to use Natural Language Processing (NLP).

Also, read: Credit Card Fraud detection using Machine Learning in Python

Importing Libraries

In Machine learning using Python the libraries have to be imported like Numpy, Seaborn and Pandas.

import numpy as np
import pandas as pd 
import seaborn as sns
import re
import nltk
from nltk.corpus import stopwords
from nltk.stem.porter import PorterStemmer
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.model_selection import train_test_split
from sklearn.naive_bayes import MultinomialNB
from sklearn.metrics import f1_score

The Dataset:

Here is the link to the Datasets: test.csv, train.csv

data_train = pd.read_csv("train.csv")
print("Data shape = ",data_train.shape)
data_train.head()

Output:

id

keyword

location

text

target

0

1

NaN

NaN

Our Deeds are the Reason of this #earthquake M...

1

1

4

NaN

NaN

Forest fire near La Ronge Sask. Canada

1

2

5

NaN

NaN

All residents asked to 'shelter in place' are ...

1

3

6

NaN

NaN

13,000 people receive #wildfires evacuation or...

1

4

7

NaN

NaN

Just got sent this photo from Ruby #Alaska as ...

1

Dropping the not required columns:

data_train = data_train.drop(['location','keyword'], axis=1)
print("location and keyword columns droped successfully")
location and keyword columns droped successfully

data_train = data_train.drop('id', axis=1)
print("id column droped successfully")


data_train.columns

Output: 

Out[7]:

0    Our Deeds are the Reason of this #earthquake M...
1               Forest fire near La Ronge Sask. Canada
2    All residents asked to 'shelter in place' are ...
3    13,000 people receive #wildfires evacuation or...
4    Just got sent this photo from Ruby #Alaska as ...
5    #RockyFire Update => California Hwy. 20 closed...
6    #flood #disaster Heavy rain causes flash flood...
7    I'm on top of the hill and I can see a fire in...
8    There's an emergency evacuation happening now ...
9    I'm afraid that the tornado is coming to our a...
Name: text, dtype: object

Create corpus a feature of NLP:

corpus  = []
pstem = PorterStemmer()
for i in range(data_train['text'].shape[0]):
    #Remove unwanted words
    tweet = re.sub("[^a-zA-Z]", ' ', data_train['text'][i])
    #Transform words to lowercase
    tweet = tweet.lower()
    tweet = tweet.split()
    #Remove stopwords then Stemming it
    tweet = [pstem.stem(word) for word in tweet if not word in set(stopwords.words('english'))]
    tweet = ' '.join(tweet)
    #Append cleaned tweet to corpus
    corpus.append(tweet)
    
print("Corpus created successfully")

#Create our dictionary 
uniqueWordFrequents = {}
for tweet in corpus:
    for word in tweet.split():
        if(word in uniqueWordFrequents.keys()):
            uniqueWordFrequents[word] += 1
        else:
            uniqueWordFrequents[word] = 1
            
#Convert dictionary to dataFrame
uniqueWordFrequents = pd.DataFrame.from_dict(uniqueWordFrequents,orient='index',columns=['Word Frequent'])
uniqueWordFrequents.sort_values(by=['Word Frequent'], inplace=True, ascending=False)
uniqueWordFrequents.head(10)

Output:

216

uniqueWordFrequents['Word Frequent'].unique()

Output:

array([4746, 4721, 411, 363, 344, 311, 239, 228, 220, 216, 213, 210, 209, 201, 183, 181, 180, 178, 175, 169, 166, 164, 162, 156, 155, 153, 151, 145, 144, 143, 137, 133, 132, 131, 130, 129, 128, 125, 124, 123, 122, 121, 120, 119, 118, 117, 116, 114, 111, 110, 109, 108, 106, 105, 104, 103, 102, 101, 100, 99, 98, 97, 96, 95, 94, 93, 91, 90, 89, 88, 87, 86, 84, 83, 82, 79, 78, 77, 76, 75, 74, 73, 72, 71, 70, 69, 68, 67, 66, 65, 64, 63, 62, 61, 60, 59, 58, 57, 56, 55, 54, 53, 52, 51, 50, 49, 48, 47, 46, 45, 44, 43, 42, 41, 40, 39, 38, 37, 36, 35, 34, 33, 32, 31, 30, 29, 28, 27, 26, 25, 24, 23, 22, 21, 20, 19, 18, 17, 16, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1])

uniqueWordFrequents = uniqueWordFrequents[uniqueWordFrequents['Word Frequent'] >= 20]
print(uniqueWordFrequents.shape)
uniqueWordFrequents

Output:

(787, 1)

20

counVec = CountVectorizer(max_features = uniqueWordFrequents.shape[0])
bagOfWords = counVec.fit_transform(corpus).toarray()

X = bagOfWords
y = data_train['target']
print("X shape = ",X.shape)
print("y shape = ",y.shape)

X_train , X_test , y_train , y_test = train_test_split(X,y,test_size=0.20, random_state=55, shuffle =True)
print('data splitting successfully')

Output:
X shape =  (7613, 787)
y shape =  (7613,)
data splitting successfully

multinomialNBModel = MultinomialNB(alpha=0.1)
multinomialNBModel.fit(X_train,y_train)

print("multinomialNB model run successfully")

#evaluation Details
models = [multinomialNBModel]

for model in models:
    print(type(model).__name__,' Train Score is   : ' ,model.score(X_train, y_train))
    print(type(model).__name__,' Test Score is    : ' ,model.score(X_test, y_test))
    
    y_pred = model.predict(X_test)
    print(type(model).__name__,' F1 Score is      : ' ,f1_score(y_test,y_pred))
    print('--------------------------------------------------------------------------')

Output:
MultinomialNB  Train Score is   :  0.8022988505747126
MultinomialNB  Test Score is    :  0.7734734077478661
MultinomialNB  F1 Score is      :  0.7165160230073953
--------------------------------------------------------------------------