A Survey on Role of Machine Learning and NLP in Fake News Detection on Social Media

Abstract

Due to the enormous and exponential advancement in the online social network, the triad of Facebook, Twitter and Whatsapp posed a great challenge in the form of fake news in front of us. In recent years many events like false propaganda of the ‘US presidential election’, opinion spamming in ‘Brexit referendum’, and long-tail series of viral rumors after many natural calamities around the world, created a lot of chaos and law and order problem. Simultaneously, this rapid explosion of fake news also attracted the attention of different researchers to investigate the real cause of it and thus to developed some tools and techniques to relieve and discover the Rumors across online media as soon as possible. In this regard, the Machine Learning (ML) algorithms and Natural Language Processing (NLP) algorithms emerged as the remarkably vital and essential tool to detect fake news in the current age. NLP when aided with machine learning produced many remarkable results that were possible just by manual fact-checking or by normal text detection process. We have systematically discussed the role of NLP and machine learning in the fake news detection process, and various detection techniques based on these. Basic terminology of NLP and machine learning too explained in brief. At last, we gave light on the future trends, open issues, challenges, and potential research oriented toward NLP and ML-based approaches.