Classification Methods for Hate Speech Diffusion: Detecting the Spread of Hate Speech on Twitter

In this paper, we investigate predictive models to detect the spread of hate speech on Twitter based on diffusion patterns. We experiment with a dataset of 10,000 tweets manually labelled as hate speech or not and show that classification based solely on the sharing graph yields strong F1 scores for our task and high hate speech detection precision. We also highlight the vulnerability of existing textual hate speech detection methods to adversarial attacks and demonstrate that while our methods do not outperform stateof-the-art text models, graph-based models provide robust detection mechanisms and are able to detect instances of hate speech that missed by text classifiers. We find that graph convolutional networks produce the strongest hate speech F1 score of 0.58 and that kernel methods offer strong predictive potential. Finally, we also consider the effects of automated bots in the diffusion of hate speech content and conclude that their sharing behavior plays an insignificant role in our experiments.