Exposing Paid Opinion Manipulation Trolls

Proceedings of Recent Advances in Natural Language Processing, pages 443–450, Hissar, Bulgaria, Sep 7–9 2015. Exposing Paid Opinion Manipulation Trolls Todor Mihaylov, Ivan Koychev FMI Sofia University tbmihailov@gmail.com koychev@fmi.uni-sofia.bg Georgi D. Georgiev Ontotext AD Sofia, Bulgaria georgiev@ontotext.com Preslav Nakov Qatar Computing Research Institute HBKU, Qatar pnakov@qf.org.qa Abstract Recently, Web forums have been invaded by opinion manipulation trolls. Some trolls try to influence the other users driven by their own convictions, while in other cases they can be organized and paid, e.g., by a political party or a PR agency that gives them specific instructions what to write. Finding paid trolls automatically using machine learning is a hard task, as there is no enough training data to train a classifier; yet some test data is possi- ble to obtain, as these trolls are sometimes caught and widely exposed. In this paper, we solve the training data problem by as- suming that a user who is called a troll by several different people is likely to be such, and one who has never been called a troll is unlikely to be such. We compare the profiles of (i) paid trolls vs. (ii) “men- tioned” trolls vs. (iii) non-trolls, and we further show that a classifier trained to dis- tinguish (ii) from (iii) does quite well also at telling apart (i) from (iii). 1 Introduction During the 2013-2014 Bulgarian protests against the Oresharski cabinet, social networks and news community forums became the main “battle grounds” between supporters and opponents of the government. In that period, there was notable cen- sorship in the media, and many people who lived outside the capital did not really know what was actually happening. Moreover, there was a very notable presence of government supporters in Web forums. In series of leaked documents in the in- dependent Bulgarian media Bivol,1 it was alleged that the ruling Socialist party was paying Internet trolls with EU Parliament money. 1https://bivol.bg/en/category/b-files-en/ b-files-trolls-en The Bivol’s leaked documents revealed for the first time such a practice by a political party de- spite the problem with opinion manipulation be- ing generally notable across Eastern Europe. The reputation management documents described the following services:“Monthly posting online of 250 comments by virtual users with varied, typical and evolving profiles from different (non-recurring) IP addresses to inform, promote, balance or counter- act. The intensity of the provided online presence will be adequately distributed and will correspond to the political situation in the country.” The practice of using Internet trolls for opinion manipulation has been reality since the rise of In- ternet and community forums. It has been shown that user opinions about products, companies and politics can be influenced by opinions posted by other online users (Dellarocas, 2006). This makes it easy for companies and political parties to gain popularity by paying for “reputation management” to people that write in discussion forums and so- cial networks fake opinions from fake profiles. Yet, over time, forum users developed sensitivity about trolls, and started publicly exposing them. 2 Related Work A …

Content truncated to 500 words