Ontology-based rumor identification system using social network data

With the tremendous expansion of online social networks (OSNs) such as Twitter and Facebook, have become not only one of the utmost effective modes for viral advertising. But also, the main source of news among people nowadays. Besides these encouraging features, however, comes the risk of rumor dissemination. Due to the enormous users of OSN, there is a considerable challenge to proficiently contain the viral spread of rumors. In addition, because of a huge number of news on OSN, the manual detection of rumors is not possible. So, we need an auto-detection and authentic knowledge sharing method in a machine-readable as well as the human-readable format for OSN. In this regard, several researchers, doing research in this domain. But no method deals with uniform, authentic, and machine-readable as well as human-readable knowledge sharing among OSN users. To address this issue, ontology-based systems can be used because ontology-based systems deal with all such kinds of issues very effectively. However, for effective categorization of OSN posts, ontology-based systems should consider roles, rating, confidence level, location, and post categorization while decision making about OSN posts. Whereas, several ontology-based methods presented for rumors verification, consider only a few requirements resulting in low accuracy of authenticated posts. Experimental results verify that the proposed ontology-based approach can produce accurate results according to human understanding.