Evolution of topics and hate speech in retweet network communities

Abstract

Twitter data exhibits several dimensions worth exploring: a network dimension in the form of links between the users, textual content of the tweets posted, and a temporal dimension as the time-stamped sequence of tweets and their retweets. In the paper, we combine analyses along all three dimensions: temporal evolution of retweet networks and communities, contents in terms of hate speech, and discussion topics. We apply the methods to a comprehensive set of all Slovenian tweets collected in the years 2018–2020. We find that politics and ideology are the prevailing topics despite the emergence of the Covid-19 pandemic. These two topics also attract the highest proportion of unacceptable tweets. Through time, the membership of retweet communities changes, but their topic distribution remains remarkably stable. Some retweet communities are strongly linked by external retweet influence and form super-communities. The super-community membership closely corresponds to the topic distribution: communities from the same super-community are very similar by the topic distribution, and communities from different super-communities are quite different in terms of discussion topics. However, we also find that even communities from the same super-community differ considerably in the proportion of unacceptable tweets they post.

Retweet communities reveal the main sources of hate speech

“…This is also consistent with the very low fraction of unacceptable tweets posted by the Sports community in Fig 3 . The distribution of topics within the detected communities, the levels of topic-related hate speech, and the evolution through time are some of the interesting results reported in [ 21 ].…”

“…Several observations can be made. First, the prevailing Twitter activities are mostly biased towards political and ideological discussions, even during the emergence of the Covid-19 pandemic [ 21 ]. There is only one, relatively small, non-political community, Sports.…”

“…Another dimension of hate speech analysis are the topics which are discussed. The results of topic detection on Slovenian Twitter show that political and ideological discussions are prevailing, accounting for almost 45% of all the tweets [ 21 ]. The sports-related topic, for example, is subject of only about 12% of all the tweets, and 90% of them are acceptable.…”

“…The work presented here is an extension of our previous research in the area of evolution of retweet communities [ 20 ]. The results, obtained on the same Twitter dataset as used here, show that the Slovenian tweetosphere is dominated by politics and ideology [ 21 ], that the left-leaning communities are larger, but that the right-leaning communities and users exhibit significantly higher impact. Furthermore, we empirically show that retweet networks change relatively gradually, despite significant external events, such as the emergence of the Covid-19 pandemic and the change of government.…”

Handling Disagreement in Hate Speech Modelling

“…The Italian model was used in a work investigating relationships between hate speech and misinformation sources on the Italian YouTube [4]. The Slovenian model was used to perform an analysis on the evolution of retweet communities, hate speech and topics on the Slovenian Twitter during 2018-2020 [8][9][10].…”