Fake News Detector on Social Media Using Machine Learning

Background: Social media for news consumption is a critical sword. On the one hand, its low cost, easy access, and rapid publishing of information lead people to seek out and consume news from social media. On the other hand, it enables the wide spread of ‘fake news’, i.e., low quality news with intentionally false information. For example, 62 percent of U.S. adults get news on social media in 2016, while in 2012, only 49 percent reported seeing news on social media [1]. It was also found that social media now defeat television as the major news source [2]. It was estimated that over 1 million tweets are related to fake news ‘Pizzagate’ [3] by the end of the presidential election.

Objective: The fake news detection problem is formulated credibility score inference problem, and it aims at learning a prediction model to infer the credibility labels of news articles, creators and subjects together.

Methodology: The methodology used in this research consists of Data Pre-processing, ICF++ Method, POS Tagging, FP Growth and some creation of transection files etc.

Result and Discussion: The author of [4] have classified every tweet/post as binary classification problem. The classification is purely on the basis of source of the post/tweet. After that the results shows 15% fake, 45% real tweets & rests posts where undecided.

Conclusion & Future Work: There is evident success in detection of fake news and posts using various Machine learning approaches. However, everchanging characteristics and features of fake news in social media networks is posing a challenge in categorization of fake news. However, the main characteristic feature of deep learning is to compute hierarchical features.