Propagation Tree Is Not Deep: Adaptive Graph Contrastive Learning Approach for Rumor Detection

TL;DR

This paper introduces RAGCL, an adaptive graph contrastive learning method that effectively detects rumors by focusing on the wide, shallow structures of rumor propagation trees through principled augmentation techniques.

Contribution

It reveals the wide structure of rumor propagation trees and proposes a novel adaptive augmentation strategy for graph contrastive learning tailored for rumor detection.

Findings

RAGCL outperforms state-of-the-art methods on four datasets.

Adaptive augmentation improves rumor representation robustness.

Wide-structure RPTs are common in real-world rumor propagation.

Abstract

Rumor detection on social media has become increasingly important. Most existing graph-based models presume rumor propagation trees (RPTs) have deep structures and learn sequential stance features along branches. However, through statistical analysis on real-world datasets, we find RPTs exhibit wide structures, with most nodes being shallow 1-level replies. To focus learning on intensive substructures, we propose Rumor Adaptive Graph Contrastive Learning (RAGCL) method with adaptive view augmentation guided by node centralities. We summarize three principles for RPT augmentation: 1) exempt root nodes, 2) retain deep reply nodes, 3) preserve lower-level nodes in deep sections. We employ node dropping, attribute masking and edge dropping with probabilities from centrality-based importance scores to generate views. A graph contrastive objective then learns robust rumor representations. Extensive experiments on four benchmark datasets demonstrate RAGCL outperforms state-of-the-art methods. Our work reveals the wide-structure nature of RPTs and contributes an effective graph contrastive learning approach tailored for rumor detection through principled adaptive augmentation. The proposed principles and augmentation techniques can potentially benefit other applications involving tree-structured graphs.