Has the online discussion been manipulated? quantifying online discussion authenticity within online social media

Aviad Elyashar, Jorge Bendahan, Rami Puzis

arXiv preprint arXiv:1708.02763, 2017

Online social media (OSM) has a enormous influence in today’s world. Some individuals view OSM as fertile ground for abuse and use it to disseminate misinformation and political propaganda, slander competitors, and spread spam. The crowdturfing industry employs large numbers of bots and human workers to manipulate OSM and misrepresent public opinion. The detection of online discussion topics manipulated by OSM emph{abusers} is an emerging issue attracting significant attention. In this paper, we propose an approach for quantifying the authenticity of online discussions based on the similarity of OSM accounts participating in the discussion to known abusers and legitimate accounts. Our method uses several similarity functions for the analysis and classification of OSM accounts. The proposed methods are demonstrated using Twitter data collected for this study and previously published emph{Arabic honeypot dataset}. The former includes manually labeled accounts and abusers who participated in crowdturfing platforms. Evaluation of the topic’s authenticity, derived from account similarity functions, shows that the suggested approach is effective for discriminating between topics that were strongly promoted by abusers and topics that attracted authentic public interest.