Integrating Semantic and Structural Information with Graph Convolutional Network for Controversy Detection

TL;DR

This paper introduces TPC-GCN and DTPC-GCN models that effectively combine semantic content and reply structure to improve controversy detection on social media, demonstrating superior performance and generalizability.

Contribution

The paper proposes novel graph convolutional network models that integrate semantic and structural information for controversy detection, addressing limitations of previous methods.

Findings

Models outperform existing controversy detection methods.

Effective integration of semantic and structural information.

Models generalize well to different topics.

Abstract

Identifying controversial posts on social media is a fundamental task for mining public sentiment, assessing the influence of events, and alleviating the polarized views. However, existing methods fail to 1) effectively incorporate the semantic information from content-related posts; 2) preserve the structural information for reply relationship modeling; 3) properly handle posts from topics dissimilar to those in the training set. To overcome the first two limitations, we propose Topic-Post-Comment Graph Convolutional Network (TPC-GCN), which integrates the information from the graph structure and content of topics, posts, and comments for post-level controversy detection. As to the third limitation, we extend our model to Disentangled TPC-GCN (DTPC-GCN), to disentangle topic-related and topic-unrelated features and then fuse dynamically. Extensive experiments on two real-world datasets demonstrate that our models outperform existing methods. Analysis of the results and cases proves that our models can integrate both semantic and structural information with significant generalizability.