Online Library

The dramatic growth of the Web has motivated researchers to extract knowledge from enormous repositories and to exploit the knowledge in myriad applications. In this study, we focus on natural language processing (NLP) and, more concretely, the emerging field of affective computing to explore the automation of understanding human emotions from texts. This paper continues previous efforts to utilize and adapt affective techniques into different areas to gain new insights. This paper proposes two novel feature extraction methods that use the previous sentic computing resources AffectiveSpace and SenticNet. These methods are efficient approaches for extracting affect-aware representations from text. In addition, this paper presents a machine learning framework using an ensemble of different features to improve the overall classification performance. Following the description of this approach, we also study the effects of known feature extraction methods such as TF-IDF and SIMilarity-based sentiment projectiON (SIMON). We perform a thorough evaluation of the proposed features across five different datasets that cover radicalization and hate speech detection tasks. To compare the different approaches fairly, we conducted a statistical test that ranks the studied methods. The obtained results indicate that combining affect-aware features with the studied textual representations effectively improves performance. We also propose a criterion considering both classification performance and computational complexity to select among the different methods.

This briefing is the first in ISD’s Gaming and Extremism Series exploring the role online gaming plays in the strategy of far-right extremists in the UK and around the world. This is part of a broader programme on the “Future of Extremism” being delivered by ISD in the second half of 2021, charting the transformational shifts in the extremist threat landscape two decades since 9/11, and the policy strategies required to counter the next generation of extremist threats. The series will include reports exploring extremist activity on four gaming-related platforms: Steam, Discord, DLive and Twitch.

This study explores the internal dynamics and networks of terrorist groups in cyberspace—in particular, Al-Qaeda and the Islamic State of Iraq and al-Sham (ISIS). Using a “Global Cyberterrorism Dataset” that features data on cyberterror attacks between 2011 and 2016, this research analyzes these two terrorist groups through the lens of a cyber-conflict theory that integrates conflict theory with Jaishankar’s space transition theory. Through a network analysis methodology, we examine the invisible relationships and connections between the national origins and target countries of cyberterror attacks. The analysis focuses on the networks of national origins of terrorists and victims, network structures of Al-Qaeda and ISIS actors, and clustering networks of Al-Qaeda and ISIS cyberterrorists. Results indicate that terror in cyberspace is ubiquitous, more flexible than traditional terrorism, and that cyberattacks mostly occurred within the countries of origin. We conclude by discussing the complex features of cyberterror networks and identify some of the geostrategic implications of the divergent cyber strategies adopted by Al-Qaeda and ISIS.

Hate has been a growing concern with hate-groups and individuals using the Internet, or more specifically, social media platforms, to globalize hate. Since these social media platforms can connect users around the world, hate-organizations are using these connections as opportunities to recruit candidates and spread their propaganda. Without opposing views, these extreme viewpoints can establish themselves as legitimate and then be used to incite hate in individuals. Thus, these extreme viewpoints must be countered by similar messages to discourage this online hate, and one such way is to use the same platforms through grassroots movements. This paper presents a case study which was conducted on a class of Criminology students who implemented a grassroots community-based campaign called Unity Starts with U (USwithU) to counter-hate in a community by using social media platforms to spread messages of inclusion and share experiences. The results from the campaign showed improvements on people’s attitude towards hate at the local community level. Based on literature and this campaign, policy recommendations are suggested for policymakers to consider when creating or making improvements on counter-narrative programs.

This paper analyses the impact of a series of mass shootings committed in 2018–2019 by right-wing extremists on 8chan/pol, a prominent far-right online forum. Using computational methods, it offers a detailed examination of how attacks trigger shifts in both forum activity and content. We find that while each shooting is discussed by forum participants, their respective impact varies considerably. We highlight, in particular, a ‘Tarrant effect’: the considerable effect Brenton Tarrant’s attack of two mosques in Christchurch, New Zealand, had on the forum. Considering the rise in far-right terrorism and the growing and diversifying online far-right ecosystem, such interactive offline-online effects warrant the attention of scholars and security professionals.

This article introduces the method of podcast ethnography. The method encompasses three general stages: to explore a podcast from a particular social field, to engage with it through careful, ethnographic reflexivity and to examine the podcast by developing typologies and themes expedient for analysis. Podcast ethnography is beneficial due to its spatial and temporal flexibility; observing a podcast universe can be performed on the move and in parallel with other tasks. This advantage enables a much-needed breathing space for researchers inquiring vehement milieus, such as white radical nationalism. The article uses an example from this precise milieu in Sweden – the podcast Motgift [Antidote] – to illustrate and flesh out the potentials and challenges of applying the method’s three stages. In so doing, the article argues for inclusion of podcast ethnography into the extended family of ethnographic methods.