Fake News Detection and Prevention Using Artificial Intelligence Techniques: A Review of a Decade of Research

Abstract

The circulation of fake news among people is not something new, as it has been present ages ago. In a connected world, due to the rapid development in the means of communication, fake news has become a very dangerous factor in daily life due to its massive impact. Furthermore, the size and speed of data shared through mediums makes it is difficult to differentiate fake and legitimate information. Social media allows sharing of data with low cost and easy access. This causes a harmful impact on individuals and society. Fake news classification and related topics has become an attractive topic for researchers in many disciplines such as journalism, natural language processing and national security . This paper reviews the various methods and techniques used in solving fake news problem and investigates weaknesses in the methods and techniques used in literature review. The challenge is to find the most useful technology for detecting and mapping fake news. We concluded that many techniques - systems were designed and implemented to automate the process of detecting fake and misleading news, and also identified that deep learning techniques have a great ability to categorize and identify hidden correlation between multiple features in several fake news benchmark datasets in a way that overcomes human capabilities.