Context Matters: Incorporating Target Awareness in Conversational Abusive Language Detection

2025-08-18

AI Summary

This work addresses the neglect of conversational context in abusive language detection by investigating the role of parent tweets (i.e., original posts being replied to) in classifying the abusiveness of reply tweets. We propose a multi-dimensional contextual feature fusion framework and systematically compare the efficacy of content-based features—such as semantic similarity and sentiment consistency—with user-based features—including historical behavior and follower count. Experiments are conducted on a manually annotated dataset of parent–reply tweet pairs, using four classification models. Results show that incorporating parent-tweet content features significantly improves detection performance (up to +5.2% F1), with semantic association features yielding the largest gains; content features consistently outperform user features in discriminative power. Crucially, this study provides the first empirical evidence that target awareness—the semantic grounding of a reply in its parent tweet—is a key mechanism underlying abusive language identification. Our findings establish a novel, context-aware paradigm for abusive language detection.

Technology Category

• Natural Language Processing: Safety and Robustness
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Machine Learning: Feature Construction/Reformulation

Application Category

• Web Mining and Content Analysis: Sentiment analysis and opinion mining
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search

Abstract

Abusive language detection has become an increasingly important task as a means to tackle this type of harmful content in social media. There has been a substantial body of research developing models for determining if a social media post is abusive or not; however, this research has primarily focused on exploiting social media posts individually, overlooking additional context that can be derived from surrounding posts. In this study, we look at conversational exchanges, where a user replies to an earlier post by another user (the parent tweet). We ask: does leveraging context from the parent tweet help determine if a reply post is abusive or not, and what are the features that contribute the most? We study a range of content-based and account-based features derived from the context, and compare this to the more widely studied approach of only looking at the features from the reply tweet. For a more generalizable study, we test four different classification models on a dataset made of conversational exchanges (parent-reply tweet pairs) with replies labeled as abusive or not. Our experiments show that incorporating contextual features leads to substantial improvements compared to the use of features derived from the reply tweet only, confirming the importance of leveraging context. We observe that, among the features under study, it is especially the content-based features (what is being posted) that contribute to the classification performance rather than account-based features (who is posting it). While using content-based features, it is best to combine a range of different features to ensure improved performance over being more selective and using fewer features. Our study provides insights into the development of contextualized abusive language detection models in realistic settings involving conversations.

Problem

• Detecting abusive language in conversational social media contexts
• Evaluating impact of parent tweet context on abuse detection
• Comparing content-based vs account-based contextual features

Innovation

• Leveraging parent tweet context for detection
• Combining content-based and account-based features
• Testing four classification models for generalization