Problem Statement

Development of an Algorithm for Fake News Classifier

Need: Now-a-days due to advent of social media and other social and news delivery platforms, a lot of news which quickly becomes viral, based on the no. of views. However the question of authenticity of these news, gets raised. Even a small conflicting news which isn’t true many times can lead to false bases for many debates and human mindset. So by using the Natural Language Processing technique, an effort is made to validate such viral news.

Introduction

The rise of fake news during the 2016 U.S. Presidential Election highlighted not only the dangers of the effects of fake news but also the challenges presented when attempting to separate fake news from real news. Fake news may be a relatively new term but it is not necessarily a new phenomenon. Fake news has technically been around at least since the appearance and popularity of one-sided, partisan newspapers in the 19th century. However, advances in technology and the spread of news through different types of media have increased the spread of fake news today.

Existing Techniques

Fake And Real News Dataset

Classifying the news

Can you use this dataset to make an algorithm that is able to determine if an article is fake news or not?

Here is a link to the dataset

Model Pipeline

Stemming

Output of Tokenization

Results

Conclusion

From different ML models used, we conclude that SVM and Logistic regression provide higher accuracy, compared to Naive Bayes used.