Online Library

From its start, the so-called Islamic State of Iraq and the Levant (ISIL/ISIS) has been successfully exploiting social media networks, most notoriously Twitter, to promote its propaganda and recruit new members, resulting in thousands of social media users adopting pro-ISIS stance every year. Automatic identification of pro-ISIS users on social media has, thus, become the centre of interest for various governmental and research organisations. In this paper we propose a semantic-based approach for radicalisation detection on Twitter. Unlike most previous works, which mainly rely on the lexical and contextual representation of the content published by Twitter users, our approach extracts and makes use of the underlying semantics of words exhibited by these users to identify their pro/anti-ISIS stances. Our results show that classifiers trained from words’ semantics outperform those trained from lexical and network features by 2% on average F1-measure.

Compared to the more spectacular elements of its media repertoire—the slick recruitment campaigns on social media, the artfully composed battlefield footage, the grisly executions—I.S.I.S.’s infographics may seem dull, even trivial. Indeed, these data visualizations have gone largely unremarked, eliciting more bemusement than serious consideration. Against the tendency to discount these images, however, I argue that when I.S.I.S. turns toward charts and diagrams to represent its operations, it launches a stealthy but substantial epistemological challenge to media outlets that depict it as backward and irrational and rely on command of information as an index of Western power. Comparing infographics produced about I.S.I.S. and those produced by the group, I demonstrate that, despite their obvious differences, both types of infographics evince common preoccupations. Like Western news sources, I.S.I.S. creates infographics to map attacks, plot territorial gains, tally and categorize casualties, and track the types of weapons deployed. News media and I.S.I.S. infographics diverge primarily in their affective resonance, as similar information signifies in radically different ways. Ultimately, by producing and circulating these infographics, I.S.I.S. renders simultaneously renders itself more and less intelligible to outsiders: encapsulating its story while confounding prevailing representations as it weaponizes information.

VOX-Pol Blog post.

The internet permeates all aspects of modern life and violent extremism is no exception. Although the level of importance is sometimes disputed, few would deny the role that online communication has in driving people towards violent extremist groups. While there are various perspectives on the exact nature of this process, it is increasingly agreed upon that it is rare for individuals to radicalise entirely in absence of any outside communication. Radicalisation remains a social phenomenon and the fact that some of these social interactions have migrated online does not change this. Extremists do not simply produce and disseminate propaganda and then move straight to offline recruitment, they utilise peer to peer messaging applications contained within social media platforms to engage in direct personal contact with potential recruits to their cause. Sometimes these online conversations completely displace offline recruitment.

Extremist propaganda is often removed and there are examples of nascent efforts to counter this throught the creation of counter-narrative campaigns. Counter-narratives, and offline counter-recruitment programmes such as EXIT and Channel, counter efforts of extremists to promote propaganda online and recruit in the offline world. This highlights the fact that there is a crucial piece missing in our efforts to counter recruitment to extremist groups; the proactive utilisation of peer to peer messaging systems online to engage with those expressing extremist sympathies.

Over the course of ISD’s management of the AVE network we encountered a number of isolated attempts to engage directly with extremists online, from former extremists infiltrating extremist forums to Twitter conversations between activists and extremist sympathisers. However none of these efforts had been attempted at scale none had had testing and success
metrics built in from the start. As such, any evidence gained as to their effectiveness was anecdotal at best.

This brief analyses the impact of the Christchurch Call to Action, issued to gather countries and technology companies to stop the use of the internet for disseminating violent extremist content. The Call was the result of a summit organised shortly after a terrorist attack in New Zealand in March 2019. This brief finds that the Call lacks clear conceptual definitions and is singularly focused on social media platforms. It also raises questions about how such efforts can maintain a balance between safeguarding digital platforms against terrorist and violent content, while preserving essential freedoms including of speech and expression.

VOX-Pol Blog Post.