Online Library

With the spread of social networks and their unfortunate use for hate speech, automatic detection of the latter has become a pressing problem. In this paper, we reproduce seven state-of-the-art hate speech detection models from prior work, and show that they perform well only when tested on the same type of data they were trained on. Based on these results, we argue that for successful hate speech detection, model architecture is less important than the type of data and labeling criteria. We further show that all proposed detection techniques are brittle against adversaries who can (automatically) insert typos, change word boundaries or add innocuous words to the original hate speech. A combination of these methods is also effective against Google Perspective – a cutting edge solution from industry. Our experiments demonstrate that adversarial training does not completely mitigate the attacks, and using character-level features makes the models systematically more attack-resistant than using word-level features.

With the spread of social networks and their unfortunate use for hate speech, automatic detection of the latter has become a pressing problem. In this paper, we reproduce seven state-of-the-art hate speech detection models from prior work, and show that they perform well only when tested on the same type of data they were trained on. Based on these results, we argue that for successful hate speech detection, model architecture is less important than the type of data and labeling criteria. We further show that all proposed detection techniques are brittle against adversaries who can (automatically) insert typos, change word boundaries or add innocuous words to the original hate speech. A combination of these methods is also effective against Google Perspective – a cuttingedge solution from industry. Our experiments demonstrate that adversarial training does not completely mitigate the attacks, and using character-level features makes the models systematically more
attack-resistant than using word-level features.

VOX-Pol Blog Post.

VOX-Pol Blog post.

The centrality of Arab mass media to Al-Qaeda’s political strategy has long been evident. From spectacular terror attacks designed for maximal media exposure, to carefully timed videos from Osama bin Laden and his lieutenant, Ayman al- Zawahiri, to the burgeoning realm of jihadi Internet forums, Al-Qaeda the organisation has increasingly become indistinguishable from Al-Qaeda the media phenomenon. This article explores the nature of Al-Qaeda’s relationship with the Arab media, which has been poorly understood leading to wrong policy conclusions.