Jointly modelling the evolution of community structure and language in online extremist groups

2024-09-28

arXiv.org

AI Summary

This work addresses the challenge of jointly modeling the co-evolution of social structure and language in online extremist communities (e.g., the anti-feminist “manosphere”). We propose the first cross-modal framework integrating temporal graph neural networks with dynamic word embeddings. Methodologically, it employs a joint optimization objective to align user and word representations temporally, enabling sequential user modeling and event-aware semantic tracking. Our key contribution lies in overcoming the limitations of static embeddings and isolated network analyses by enabling the first synchronous, time-resolved modeling of community structure and linguistic representation—along with dynamic attribution of behavioral and semantic shifts. Evaluated on manosphere data, our model significantly outperforms unimodal baselines: it accurately predicts user group affiliations, captures event-triggered lexical semantic drift, and quantifies differential propensities toward violent discourse across subcommunities.

Technology Category

• Natural Language Processing: Language Grounding & Multi-modal NLP
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Application Domains: Humanities & Computational Social Science

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Group interactions take place within a particular socio-temporal context, which should be taken into account when modelling communities. We propose a method for jointly modelling community structure and language over time, and apply it in the context of extremist anti-women online groups (collectively known as the manosphere). Our model derives temporally grounded embeddings for words and users, which evolve over the training window. We show that this approach outperforms prior models which lacked one of these components (i.e. not incorporating social structure, or using static word embeddings). Using these embeddings, we investigate the evolution of users and words within these communities in three ways: (i) we model a user as a sequence of embeddings and forecast their affinity groups beyond the training window, (ii) we illustrate how word evolution is useful in the context of temporal events, and (iii) we characterise the propensity for violent language within subgroups of the manosphere.

Problem

• Modeling social structure and language evolution jointly
• Analyzing dynamic word and user representations for clustering
• Tracking extremist groups' responses to temporal events

Innovation

• Jointly models social structure and language evolution
• Produces dynamic word and user representations
• Outperforms prior models in clustering tasks