Experimental Evaluation of Cyber Security Enabled Fake News Detection Scheme over Social Media Using Learning Principle

Users of the internet now rely almost exclusively on social networking sites to keep up with the latest developments via news. More people are exposed to breaking news on these sites in a shorter amount of time than they would be through established media outlets. When compared to more established news sources, fortunately, social networking sites have fewer safeguards in place to ensure that the information they transmit is accurate and trustworthy. Consequently, there is a requirement for automated fake news identification systems on these channels to limit the dissemination of disinformation. Conventional detection techniques from conventional news outlets are either ineffective or irrelevant when applied to the problem of detecting fraudulent information on social networking sites. In this study, we present a novel method for identifying fake news on social networking sites and halting its spread termed Learning based Fake Data Detection (LFDD). This algorithm focuses on the following: first, fake news is written with a goal to deceive readers into believing incorrect details; which renders detecting it based on news content difficult and nontrivial; thus, we need to include additional details, including user interactions on social networks, in order to aid in making an assessment. Furthermore, people’ social contacts with false news generate large amounts of details that are fragmentary, unorganized and loud, making it difficult to extract valuable data from it. This research is intended to help inquiry into the complex and timely topic of identifying false news on social networking sites.