Combining Social Network Analysis and Sentiment Analysis to Explore the Potential for Online Radicalization

You are here

This paper explores the use of crawling global social networking platforms to undercover previously unknown radicalized individuals. To prove the utility of this process the authors collect a YouTube dataset from a group that potentially has a radicalizing agenda. This dataset is analyzed using social network and sentiment analyses tools, examining the topics discussed and the sentiment polarity towards these topics. In carrying out their study, the authors look to address to two research questions: 1). Was the group population influenced by radicals who were in a position to draw others into their sphere of influence? 2). What were the differences, if any, in terms of radicalness between the content posted and the interactions engaged in between male and female group members?

This article will be of particular use to practitioners and researchers looking to use social network analysis and sentiment analysis tools to detect radicalization and extremism online. The analysis utilized in this paper applies sentiment, lexical, and social network analysis, to examine and characterize the users of radical forums. Data collection and analysis was made using multiple steps: a YouTube crawl to gather relevant data; a network analysis of this data; and lexical analysis of the corpus to inform the sentiment analysis of the documents gathered. The authors found that the group in question had no radicalizing function and was mostly devoted to religious discussion. In terms of male and female members of the groups, it was observed that women tended to be more sympathetic to politically violent actors and also less able to distinguish between the Jewish religion and the state of Israel, a division that appeared to be clearly recognized by male members. The authors noted that the higher one’s status in a group using several different measures, the less likely one is to provide information about one’s gender.

Adam Bermingham, Maura Conway, Lisa McInerney, Neil O’Hare, and Alan F. Smeaton