Academic literature on the topic 'DETECTING EXTREMIST AFFILIATIONS'

Religion is a source of mercy and peace; religious texts are one of the most critical parts of a culture's heritage, and they affect societies often in a big way; sadly, misconceptions can also make some religious people extremists. Modern social networks provide a platform for people to express themselves share their opinions and show their affiliations on many topics. This generates data in many forms like photos, videos, and texts. The authors used predefined machine learning (ML) to classify and analyze textual data from social networks. In this paper, they focus on two types of classification: religious and extremist. Extremism is independent of religious text, and therefore, they classify them separately. The work uses and compares several algorithms to classify textual data from social networks. The proposed model has achieved 93.33% accuracy for religious classification and 97% accuracy for extremism detection.

Far-right extremist communities actively promote their ideological preferences on social media. This provides researchers with opportunities to study these communities online. However, to explore these opportunities one requires a way to identify the far-right extremists’ communities in an automated way. Having analyzed the subject area of far-right extremist communities, we identified three groups of factors that influence the effectiveness of the research work. These are a group of theoretical, methodological, and instrumental factors. We developed and implemented a unique algorithm of calendar-correlation analysis (CCA) to search for specific online communities. We based CCA on a hybrid calendar correlation approach identifying potential far-right communities by characteristic changes in group activity around key dates of events that are historically crucial to those communities. The developed software module includes several functions designed to automatically search, process, and analyze social media data. In the current paper we present a process diagram showing CCA’s mechanism of operation and its relationship to elements of automated search software. Furthermore, we outline the limiting factors of the developed algorithm. The algorithm was tested on data from the Russian social network VKontakte. Two experimental data sets were formed: 259 far-right communities and the 49 most popular (not far-right) communities. In both cases, we calculated the type II error for two mutually exclusive hypotheses—far-right affiliation and no affiliation. Accordingly, for the first sample, β = 0.81. For the second sample, β = 0.02. The presented CCA algorithm was more effective at identifying far-right communities belonging to the alt-right and Nazi ideologies compared to the neo-pagan or manosphere communities. We expect that the CCA algorithm can be effectively used to identify other movements within far-right extremist communities when an appropriate foundation of expert knowledge is provided to the algorithm.

Social media is a modern person’s digital voice to project and engage with new ideas and mobilise communities—a power shared with extremists. Given the societal risks of unvetted content-moderating algorithms for Extremism , Radicalisation , and Hate speech (ERH) detection, responsible software engineering must understand the who, what, when, where, and why such models are necessary to protect user safety and free expression. Hence, we propose and examine the unique research field of ERH context mining to unify disjoint studies. Specifically, we evaluate the start-to-finish design process from socio-technical definition-building and dataset collection strategies to technical algorithm design and performance. Our 2015-2021 51-study Systematic Literature Review (SLR) provides the first cross-examination of textual, network, and visual approaches to detecting extremist affiliation, hateful content, and radicalisation towards groups and movements. We identify consensus-driven ERH definitions and propose solutions to existing ideological and geographic biases, particularly due to the lack of research in Oceania/Australasia. Our hybridised investigation on Natural Language Processing, Community Detection, and visual-text models demonstrates the dominating performance of textual transformer-based algorithms. We conclude with vital recommendations for ERH context mining researchers and propose an uptake roadmap with guidelines for researchers, industries, and governments to enable a safer cyberspace.

The main motivation behind this research paper is to address the issue of identifying extremist affiliations on social media platforms. With the rise of social media, people have been given the power to express their opinions and emotions on a global scale, which has led to the emergence of a new form of communication. Unfortunately, some individuals and organizations have been using these platforms to spread hate and propaganda, and even recruit individuals to join their extremist causes. This has created a serious threat to national and global security. Sentiment analysis, specifically opinion mining, has emerged as an important tool for identifying and tracking extremist activities on social media. The proposed deep learning model that utilizes Distil BERT algorithm aims to improve the accuracy of classification by combining text and emoticons for sentiment analysis. The model captures sentiment expressed in both text and emoticons, highlighting the significance of including emoticons in sentiment analysis. This study has the potential to contribute significantly to the field of sentiment analysis and social media monitoring, ultimately aiding in the fight against extremism. The implications of this research can be applied to sentiment analysis in social media and extended to other social media platforms that use emojis to express opinions and emotions. By identifying tweets that support or relate to extremist affiliations, the proposed model can help authorities monitor such activities on social media platforms and take appropriate actions.