User-Generated Content (UGC) Credibility on Social Media Using Sentiment Classification

Abstract

Web 2.0 technologies have seen a big evolution recently leading to the existence of a huge amount of unreliable and misleading content due to the openness and low publishing barrier nature of the content generated through social media platforms. As a fact, the User-Generated Content (UGC) on social media platforms suffers from a lack of professional gatekeepers to monitor this content. Consequently, most online users fall into the trap of being misled through fake information that spreads rapidly. They usually rely on this information without any verification and this prevents them from making accurate decisions concerning their social lives, politics, or business events. Because online users face difficulty in finding which piece of information is credible or not, the researchers found that assessing User-Generated Content (UGC) of social media is very important in resolving the issue of credibility. This paper adapted some of the existing literature and concluded that many previous approaches have investigated information credibility on Twitter and a limited number of Facebook for proposing a new approach for measuring posts credibility. The proposed model used to measure the credibility of Facebook posts through a formula combined from the page profile rank and the post-analysis score. The model was tested and achieved 87.45 % accuracy.