Social Hatred: Efficient Multimodal Detection of Hatemongers

2025-06-24

AI Summary

This study addresses the challenge of detecting implicit hate speech—such as coded language, dog whistles, and racial manipulation—in social media, where single-post analysis proves insufficient. We propose a user-level, multimodal approach to identify hate propagators by jointly modeling textual semantics (via NLP), behavioral patterns (e.g., posting frequency and topic preferences), and social network structure (using graph neural networks) within broader sociocultural contexts. Our key contribution is the first shift in hate detection granularity from *content* to *user*, enabling holistic, cross-platform generalization (across Twitter, Gab, and Parler). Experiments demonstrate significant improvements over unimodal baselines—particularly text-only or graph-only models—in precision, robustness, and implicit expression detection. The method offers a scalable, deployable framework for large-scale online hate governance.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Machine Learning: Multimodal Learning
• Application Domains: Humanities & Computational Social Science

Application Category

• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Automatic detection of online hate speech serves as a crucial step in the detoxification of the online discourse. Moreover, accurate classification can promote a better understanding of the proliferation of hate as a social phenomenon. While most prior work focus on the detection of hateful utterances, we argue that focusing on the user level is as important, albeit challenging. In this paper we consider a multimodal aggregative approach for the detection of hate-mongers, taking into account the potentially hateful texts, user activity, and the user network. Evaluating our method on three unique datasets X (Twitter), Gab, and Parler we show that processing a user's texts in her social context significantly improves the detection of hate mongers, compared to previously used text and graph-based methods. We offer comprehensive set of results obtained in different experimental settings as well as qualitative analysis of illustrative cases. Our method can be used to improve the classification of coded messages, dog-whistling, and racial gas-lighting, as well as to inform intervention measures. Moreover, we demonstrate that our multimodal approach performs well across very different content platforms and over large datasets and networks.

Problem

• Detect hate-mongers using multimodal user-level analysis
• Improve hate speech detection with social context and activity
• Evaluate method across diverse platforms (Twitter, Gab, Parler)

Innovation

• Multimodal aggregative approach for hate detection
• User-level analysis with text, activity, network
• Cross-platform performance on large datasets