Online Library

To prevent and counter the spread of illegal hate speech online, in May 2016, the Commission agreed with Facebook, Microsoft, Twitter and YouTube a “Code of conduct on countering illegal hate speech online”. The implementation of the Code of Conduct is evaluated through a regular monitoring exercise set up in collaboration with a network of organisations located in the different EU countries. Using a commonly agreed methodology, these organisations test how the IT companies are implementing the commitments in the Code.

To prevent and counter the spread of illegal hate speech online, in May 2016, the Commission agreed with Facebook, Microsoft, Twitter and YouTube a “Code of conduct on countering illegal hate speech online”. The implementation of the Code of Conduct is evaluated through a regular monitoring exercise set up in collaboration with a network of organisations located in the different EU countries. Using a commonly agreed methodology, these organisations test how the IT companies are implementing the commitments in the Code.

VOX-Pol Blog post.

This research analyses cloaked Facebook pages that are created to spread political propaganda by cloaking a user profile and imitating the identity of a political opponent in order to spark hateful and aggressive reactions. This inquiry is pursued through a multi-sited online ethnographic case study of Danish Facebook pages disguised as radical Islamist pages, which provoked racist and anti-Muslim reactions as well as negative sentiments towards refugees and immigrants in Denmark in general. Drawing on Jessie Daniels’ critical insights into cloaked websites, this research furthermore analyses the epistemological, methodological and conceptual challenges of online propaganda. It enhances our understanding of disinformation and propaganda in an increasingly interactive social media environment and contributes to a critical inquiry into social media and subversive politics.

The rise of technology has brought with it many benefits but also the potential for great dangers. In particular, Information Communication Technology (ICT) is involved in many facets of life-influencing systems, which range from power plants to airports. Terrorists are now realising the great possibilities of interfering with critical infrastructure. Remote access, reduced costs, automation, replication, speed, direct effect, varied targets and anonymity are all benefits that make attacking computers and networks in cyberspace an attractive solution. ICT could thus serve as a powerful instrument to advance political and ideological viewpoints. The ICT landscape now faces an emerging threat in the form of cyberterrorists. However, it is important not to incorrectly perceive ordinary cyber attacks as cyberterrorism. Cyberterrorism is different from cybercrime in that is has differing motives, attack goals, techniques and intended effects. The motivation for cyberterrorism largely stems from political and ideological views (religious, social activism, retributional). Cyber attacks are mainly driven by financial theft, fraud or espionage, whereas cyberterrorism aims to create publicity for a cause and leave a high impact. In this study, a Cyberterrorism Life Cycle (CLC) Model is developed in order to demonstrate the various factors that lead to the establishment and growth of cyberterrorism. The model depicts the various strategic and technical issues that are relevant to the field. Overall, this model aims to structure the dynamic interaction of the behavioural and technological factors that influence the development of cyberterrorism. Throughout the research, various factors that are influential to cyberterrorism are investigated. The research follows a systematic approach of revealing various underlying issues and thereafter compiling the holistic CLC model to depict these critical issues. Part 1 introduces cyberterrorism and provides the background to the field by discussing incidents and example groups. Initially, the concept of cyberterrorism is explored and the proposed definition tested. Part 2 looks at investigating cyberterrorism more deeply. A conceptual framework is presented that introduces the most pertinent factors in the field of cyberterrorism. Next, the traditional and innovative use of the Internet to carry out and support terrorism is explored. Then, the study addresses the determination of additional social factors using Partial Least Squares Path Modelling. In Part 3, the field of cyberterrorism is more intensely studied. Cyberterrorism is mapped to the Observe-Orient- Decide-Act (OODA) loop, which will form the basis of the CLC model. Thereafter, the most influential concepts essential to the field of cyberterrorism are applied in order to classify attacks as cyberterrorism using ontologies. Furthermore, in Part 3, countermeasures are discussed to look at ways to combat cyberterrorism. Part 4 forms the crux of the research. The CLC model is presented as a structured representation of the various influential factors relevant to cyberterrorism. Thereafter, the CLC model is simulated to show the field more dynamically. Overall, the CLC model presented in this study aims to show the interaction of the various strategic, behavioural and technical issues. The CLC model can help elucidate the reasons for attraction into extremist groups and how attacks are carried out.

Common approaches to text categorization essentially rely either on n-gram counts or on word embeddings. This presents important difficulties in highly dynamic or quickly-interacting environments, where the appearance of new words and/or varied misspellings is the norm. A paradigmatic example of this situation is abusive online behavior, with social networks and media platforms struggling to effectively combat uncommon or nonblacklisted hate words. To better deal with these issues in those fast-paced environments, we propose using the error signal
of class-based language models as input to text classification algorithms. In particular, we train a next-character prediction model for any given class, and then exploit the error of such class-based models to inform a neural network classifier. This way, we shift from the ability to describe seen documents to the ability to predict unseen content. Preliminary studies using out-of-vocabulary splits from abusive tweet data show promising results, outperforming competitive text categorization strategies by 4–11%.