Disrupting networks of hate: characterising hateful networks and removing critical nodes

Abstract

AbstractHateful individuals and groups have increasingly been using the Internet to express their ideas, spread their beliefs and recruit new members. Understanding the network characteristics of these hateful groups could help understand individuals’ exposure to hate and derive intervention strategies to mitigate the dangers of such networks by disrupting communications. This article analyses two hateful followers’ networks and three hateful retweet networks of Twitter users who post content subsequently classified by human annotators as containing hateful content. Our analysis shows similar connectivity characteristics between the hateful followers networks and likewise between the hateful retweet networks. The study shows that the hateful networks exhibit higher connectivity characteristics when compared to other “risky” networks, which can be seen as a risk in terms of the likelihood of exposure to, and propagation of, online hate. Three network performance metrics are used to quantify the hateful content exposure and contagion: giant component (GC) size, density and average shortest path. In order to efficiently identify nodes whose removal reduced the flow of hate in a network, we propose a range of structured node-removal strategies and test their effectiveness. Results show that removing users with a high degree is most effective in reducing the hateful followers network connectivity (GC, size and density), and therefore reducing the risk of exposure to cyberhate and stemming its propagation.