Finding Opinion Manipulation Trolls in News Community Forums

Proceedings of the 19th Conference on Computational Language Learning, pages 310–314, Beijing, China, July 30-31, 2015. c⃝2015 Association for Computational Linguistics Finding Opinion Manipulation Trolls in News Community Forums Todor Mihaylov FMI Sofia University tbmihailov@gmail.com Georgi D. Georgiev Ontotext AD Sofia, Bulgaria georgiev@ontotext.com Preslav Nakov Qatar Computing Research Institute HBKU, Qatar pnakov@qf.org.qa Abstract The emergence of user forums in elec- tronic news media has given rise to the proliferation of opinion manipulation trolls. Finding such trolls automatically is a hard task, as there is no easy way to recognize or even to define what they are; this also makes it hard to get training and testing data. We solve this issue pragmati- cally: we assume that a user who is called a troll by several people is likely to be one. We experiment with different variations of this definition, and in each case we show that we can train a classifier to distinguish a likely troll from a non-troll with very high accuracy, 82–95%, thanks to our rich feature set. 1 Introduction With the rise of social media, it became normal for people to read and follow other users’ opinion. This created the opportunity for corporations, gov- ernments and others to distribute rumors, misin- formation, speculation and to use other dishonest practices to manipulate user opinion (Derczynski and Bontcheva, 2014a). They could consistently use trolls (Cambria et al., 2010), write fake posts and comments in public forums, thus making ve- racity one of the challenges in digital social net- working (Derczynski and Bontcheva, 2014b). The practice of using opinion manipulation trolls has been reality since the rise of Internet and community forums. It has been shown that user opinions about products, companies and politics can be influenced by posts by other users in online forums and social networks (Dellarocas, 2006). This makes it easy for companies and political par- ties to gain popularity by paying for “reputation management” to people or companies that write in discussion forums and social networks fake opin- ions from fake profiles. In Europe, the problem has emerged in the con- text of the crisis in Ukraine.12 There have been a number of publications in news media describing the behavior of organized trolls that try to manipu- late other users’ opinion.345 Still, it is hard for fo- rum administrators to block them as trolls try not to violate the forum rules. 2 Related Work Troll detection and offensive language use are un- derstudied problems (Xu and Zhu, 2010). They have been addressed using analysis of the seman- tics and sentiment in posts to filter out trolls (Cam- bria et al., 2010); there have been also studies of general troll behavior (Herring et al., 2002; Buck- els et al., 2014). Another approach has been to use lexico-syntactic features about user’s writing style, structure and specific cyber-bullying con- tent (Chen et al., 2012); cyber-bullying was de- tected using sentiment analysis (Xu et al., 2012); graph-based approaches over signed social net- works have been used as well (Ortega et …

Content truncated to 500 words