Journal Article |
Linguistic Patterns for Code Word Resilient Hate Speech Identification
View Abstract
The permanent transition to online activity has brought with it a surge in hate speech discourse. This has prompted increased calls for automatic detection methods, most of which currently rely on a dictionary of hate speech words, and supervised classification. This approach often falls short when dealing with newer words and phrases produced by online extremist communities. These code words are used with the aim of evading automatic detection by systems. Code words are frequently used and have benign meanings in regular discourse, for instance, “skypes, googles, bing, yahoos” are all examples of words that have a hidden hate speech meaning. Such overlap presents a challenge to the traditional keyword approach of collecting data that is specific to hate speech. In this work, we first introduced a word embedding model that learns the hidden hate speech meaning of words. With this insight on code words, we developed a classifier that leverages linguistic patterns to reduce the impact of individual words. The proposed method was evaluated across three different datasets to test its generalizability. The empirical results show that the linguistic patterns approach outperforms the baselines and enables further analysis on hate speech expressions.
|
2021 |
Calderón, F.H., Balani, N., Taylor, J., Peignon, M., Huang, Y.H. and Chen, Y.S. |
View
Publisher
|
Report |
Can the Right Meme? (And How?): A Comparative Analysis of Three Online Reactionary Meme Subcultures
View Abstract
This report analyses memes propagated among three online socio‐political groups drawn from sample datasets pulled from social media sites often used by adherents of each group. These groups include those connected to the India‐based Hindutva, US‐based neo‐Nazis and those engaging in pro‐Rittenhouse communications in late 2020. The authors chose the groups based on similarities in their ideological goals, race‐based nationalism and their close association with political violence in their respective countries.
|
2021 |
Stall, H., Prasad, H. and Foran, D. |
View
Publisher
|
Report |
The Iron March Forum and the Evolution of the “Skull Mask” Neo-Fascist Network
View Abstract
The backbone of the “skull mask” transnational neo-fascist accelerationist network—whose nodes include terror groups such as Atomwaffen, the Base, and Feuerkrieg Division—is a group of organizations that grew out of Iron March, a neo-fascist web forum that was active from 2011 to 2017. The history of the Iron March network shows that violent extremist movements can develop from online communities even in the absence of a territorial base and without regular in-person contact between members. Iron March provided a closed social space where young neo-fascists who did not fit in well in established neo-fascist organizations could create a transnational collective identity. Eventually, Iron March users sought each other out in person and created local groups that remained networked together by virtue of their common origin in the community created on the web forum. The network’s transition from activism to terrorism was facilitated by the introduction of violent ritualistic initiation practices derived from the writings of the Order of Nine Angles, which helped to habituate members to violence as well as to create a sense of shared membership in a militant elite.
|
2021 |
Upchurch, H.E. |
View
Publisher
|
Journal Article |
Countering violent extremism using social media and preventing implementable strategies for Bangladesh
View Abstract
Globally, more than 85% of youth use social media daily in the medium of Facebook, Youtube, Twitter, etc., which is more than 70% for Bangladesh. The young population of Bangladesh is rapidly embracing social media through the internet and afflicted with the malaise of countering violent extremism (CVE), often through Facebook. Given the increasing connectedness that the internet and social media offer, it is crucial that the fight against CVE shift to the digital space. Extremists are increasingly adopting novel ways and means based on technology to draw unsuspecting youth to their cause. It is essential to establish effective implementable strategies to stop the CVE activities using social media in Bangladesh. This study aims to identify existing initiatives globally in the space of disruptive online technologies that have yielded some success in preventing CVE. Various publications such as journal and news articles, TV news, and blogs have been used as data sources for this study. Also, fifteen expert interviews have been conducted to identify the most effective strategies for CVE in Bangladesh. Through the content analysis, the study highlights successful efforts and explores technology-based initiatives that can be deployed in Bangladesh to minimize the impact of VE activities through online technology. Finally, recommendations for strategies to restrict VE activities through technologies have been suggested that can be potentially implemented by the Bangladesh government by coordinating with international donor agencies and CVE practitioners. The research output recommends that Bangladesh and other less developed countries can concurrently deal with CVE by successfully using cutting-edge online/digital technologies.
|
2021 |
Amit, S., Barua, L. and Kafy, A.A. |
View
Publisher
|
Journal Article |
Online Extremism Detection: A Systematic Literature Review With Emphasis on Datasets, Classification Techniques, Validation Methods, and Tools
View Abstract
Social media platforms are popular for expressing personal views, emotions and beliefs. Social media platforms are influential for propagating extremist ideologies for group-building, fund-raising, and recruitment. To monitor and control the outreach of extremists on social media, detection of extremism in social media is necessary. The existing extremism detection literature on social media is limited by specific ideology, subjective validation methods, and binary or tertiary classification. A comprehensive and comparative survey of datasets, classification techniques, validation methods with online extremism detection tool is essential. The systematic literature review methodology (PRISMA) was used. Sixty-four studies on extremism research were collected, including 31 from SCOPUS, Web of Science (WoS), ACM, IEEE, and 33 thesis, technical and analytical reports using Snowballing technique. The survey highlights the role of social media in propagating online radicalization and the need for extremism detection on social media platforms. The review concludes lack of publicly available, class-balanced, and unbiased datasets for better detection and classification of social-media extremism. Lack of validation techniques to evaluate correctness and quality of custom data sets without human interventions, was found. The information retrieval unveiled that contemporary research work is prejudiced towards ISIS ideology. We investigated that deep learning based automated extremism detection techniques outperform other techniques. The review opens the research opportunities for developing an online, publicly available automated tool for extremism data collection and detection. The survey results in conceptualization of architecture for construction of multi-ideology extremism text dataset with robust data validation techniques for multiclass classification of extremism text.
|
2021 |
Gaikwad, M., Ahirrao, S., Phansalkar, S. and Kotecha, K. |
View
Publisher
|
Report |
Human Rights Assessment: Global Internet Forum to Counter Terrorism
View Abstract
The Global Internet Forum to Counter Terrorism (GIFCT) commissioned BSR to conduct a human rights assessment of its strategy, governance, and activities. The purpose of this assessment is to identify actual and potential human rights impacts (including both risks and opportunities) arising from GIFCT’s work and make recommendations for how GIFCT and its participants can address these impacts. BSR undertook this human rights review from December 2020 to May 2021. This assessment combines human rights assessment methodology based on the UN Guiding Principles on Business and Human Rights (UNGPs) with consideration of the human rights principles, standards, and methodologies upon which the UNGPs were built. This review was funded by GIFCT, though BSR retained editorial control over its contents.
|
2021 |
Allison-Hope, D., Andersen, L. and Morgan, S. |
View
Publisher
|