Retweet communities reveal the main sources of hate speech Bojan Evkoski1,2, AndraˇzPelicon1,2, Igor Mozetiˇc1, Nikola Ljubeˇsi´c1,3 Petra Kralj Novak1 1 Department of Knowledge Technologies, Jozef Stefan Institute, Jamova 39, Ljubljana, Slovenia 2 Jozef Stefan International Postgraduate School, Ljubljana, Slovenia 3 Faculty of Information and Communication Sciences, University of Ljubljana, Slovenia *
[email protected] Abstract We address a challenging problem of identifying main sources of hate speech on Twitter. On one hand, we carefully annotate a large set of tweets for hate speech, and deploy advanced deep learning to produce high quality hate speech classification models. On the other hand, we create retweet networks, detect communities and monitor their evolution through time. This combined approach is applied to three years of Slovenian Twitter data. We report a number of interesting results. Hate speech is dominated by offensive tweets, related to political and ideological issues. The share of unacceptable tweets is moderately increasing with time, from the initial 20% to 30% by the end of 2020. Unacceptable tweets are retweeted significantly more often than acceptable tweets. About 60% of unacceptable tweets are produced by a single right-wing community of only moderate size. Institutional Twitter accounts and media accounts post significantly less unacceptable tweets than individual accounts. However, the main sources of unacceptable tweets are anonymous accounts, and accounts that were suspended or closed during the last three years. 1 Introduction Most of the current approaches to detect and characterize hate speech focus solely on the content of posts in online social networks. They do not consider the network structure, nor the roles and types of users generating and retweeting hate speech.