# Jointly modelling the evolution of social structure and language in online communities

arXiv: https://arxiv.org/abs/2409.19243 · HTML: https://mlsanity.com/paper/2409.19243/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 59 references that are in the map.
- First posted on arXiv 28 Sep 2024.

## Builds on (most cited first, 9 of 9)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [word2vec Explained: deriving Mikolov et al.'s negative-sampling word-embedding method](https://mlsanity.com/paper/1402.3722/) (2014)
- [Dynamic Word Embeddings for Evolving Semantic Discovery](https://mlsanity.com/paper/1703.00607/) (2017)
- [A Framework for the Computational Linguistic Analysis of Dehumanization](https://mlsanity.com/paper/2003.03014/) (2020)
- [Augmenting Neural Response Generation with Context-Aware Topical Attention](https://mlsanity.com/paper/1811.01063/) (2018)
- [From Dogwhistles to Bullhorns: Unveiling Coded Rhetoric with Language Models](https://mlsanity.com/paper/2305.17174/) (2023)
- [Making "fetch" happen: The influence of social and linguistic context on nonstandard word growth and decline](https://mlsanity.com/paper/1802.04140/) (2018)
- [Predicting online extremism, content adopters, and interaction reciprocity](https://mlsanity.com/paper/1605.00659/) (2016)
- [No Easy Way Out: the Effectiveness of Deplatforming an Extremist Forum to Suppress Hate and Harassment](https://mlsanity.com/paper/2304.07037/) (2023)

## Built on by (most cited first, 2 of 2)

- [Beyond Dataset Creation: Critical View of Annotation Variation and Bias Probing of a Dataset for Online Radical Content Detection](https://mlsanity.com/paper/2412.11745/) (2024)
- [IYKYK: Using language models to decode extremist cryptolects](https://mlsanity.com/paper/2506.05635/) (2025)

## Abstract (arXiv)

Group interactions take place within a particular socio-temporal context, which should be taken into account when modelling interactions in online communities. We propose a method for jointly modelling community structure and language over time. Our system produces dynamic word and user representations that can be used to cluster users, investigate thematic interests of groups, and predict group membership. We apply and evaluate our method in the context of a set of misogynistic extremist groups. Our results indicate that this approach outperforms prior models which lacked one of these components (i.e. not incorporating social structure, or using static word embeddings) when evaluated on clustering and embedding prediction tasks. Our method further enables novel types of analyses on online groups, including tracing their response to temporal events and quantifying their propensity for using violent language, which is of particular importance in the context of extremist groups.