Graph Contrastive Learning With Feature Augmentation for Rumor Detection
Bibliographic Data
| ID | 22107281 |
|---|---|
| Authors | Shaohua Li (0000-0002-0223-0796, Shanghai University of Engineering Science), Weimin Li (0000-0003-0985-0311, Shanghai University of Engineering Science), Alex Munyole Luvembe (0000-0002-2994-3446, Shanghai University of Engineering Science), Weiqin Tong (0000-0001-8300-6376, Shanghai University of Engineering Science) |
| Year | 2024 |
| Volume | 11 |
| Issue | 4 |
| Pages | 5158-5167 |
| Publication date | 2024-08-01 |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | IEEE Transactions on Computational Social Systems (JOURNAL) |
| Journal identifiers | ISSN: 2329-924X • E-ISSN: 2373-7476 |
| Publisher | Institute of Electrical and Electronics Engineers (IEEE) (PUBLISHER) |
| DOI | 10.1109/tcss.2023.3269303 |
| OpenAlex | W4372037325 |
| Language | EN |
| Citations received | 6 |
| References cited | 26 |
While online social media brings convenience to people’s communication, it has also caused the widespread spread of rumors and brought great harm. Recent deep-learning approaches attempt to identify rumors by engaging in interactive user feedback. However, the performance of these models suffers from insufficient and noisy labeled data. In this article, we propose a novel rumor detection model called graph contrastive learning with feature augmentation (FAGCL), which injects noise into the feature space and learns contrastively by constructing asymmetric structures. FAGCL takes user preference and news embedding as the initial features of the rumor propagation tree and then adopts a graph attention network to update node representations. To obtain the graph-level representation for rumor classification, FAGCL fuses multiple pooling techniques. Moreover, FAGCL adopts graph contrastive learning as an auxiliary task to constrain the representation consistency. Contrastive learning on noisy data mines the supervision information of the rumor propagation tree itself, making the model more robust and effective. Results on two real-world datasets demonstrate that our proposed FAGCL model achieves significant improvements over the baseline models
Feature learning · Graph · Linguistics · Machine learning · Natural language processing · Rumor · Complex Network Analysis Techniques · Computer Science · Misinformation and Its Impacts · Opinion Dynamics and Social Influence · Artificial Intelligence · Theoretical Computer Science
Multimodal Meta-Learning for Early Rumor Detection Based on Few-Shot Learning
Optimizing Feature Interaction via Information Bottleneck for CTR Prediction
Multipattern Integrated Networks With Contrastive Pretraining for Graph Anomaly Detection
Multiview Spatio-Temporal Learning With Dual Dynamic Graph Convolutional Networks for Rumor Detection
LLM-Augmented Contrastive Learning for Misinformation Detection in Social Networks
Rumor Detection Based on Supervised Multiprototype Contrastive Learning
| Citations per year | 3 |
|---|---|
| Citation span | 2024 - 2026 (3) |
| Highly cited | No |