CoATA: Effective Co-Augmentation of Topology and Attribute for Graph Neural Networks

2025-06-27

AI Summary

Real-world graph data often suffer from noise and incompleteness, severely degrading the performance of Graph Neural Networks (GNNs). Existing methods typically enhance graphs along a single dimension—either topology or node attributes—neglecting their intrinsic coupling. To address this, we propose CoATA, the first framework that establishes a topology–attribute dual-channel co-enhancement mechanism. CoATA constructs a bipartite graph via structural signal propagation and attribute-space projection, then applies prototype alignment and consistency-aware contrastive learning to enable mutual refinement across both spaces. By deeply integrating structural and feature correlations, CoATA significantly improves representation robustness. Extensive experiments on seven benchmark datasets demonstrate that CoATA consistently outperforms eleven state-of-the-art methods, validating the effectiveness of the proposed co-enhancement paradigm.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Bridging structured and unstructured data

Abstract

Graph Neural Networks (GNNs) have garnered substantial attention due to their remarkable capability in learning graph representations. However, real-world graphs often exhibit substantial noise and incompleteness, which severely degrades the performance of GNNs. Existing methods typically address this issue through single-dimensional augmentation, focusing either on refining topology structures or perturbing node attributes, thereby overlooking the deeper interplays between the two. To bridge this gap, this paper presents CoATA, a dual-channel GNN framework specifically designed for the Co-Augmentation of Topology and Attribute. Specifically, CoATA first propagates structural signals to enrich and denoise node attributes. Then, it projects the enhanced attribute space into a node-attribute bipartite graph for further refinement or reconstruction of the underlying structure. Subsequently, CoATA introduces contrastive learning, leveraging prototype alignment and consistency constraints, to facilitate mutual corrections between the augmented and original graphs. Finally, extensive experiments on seven benchmark datasets demonstrate that the proposed CoATA outperforms eleven state-of-the-art baseline methods, showcasing its effectiveness in capturing the synergistic relationship between topology and attributes.

Problem

• Addresses noise and incompleteness in real-world graphs
• Enhances GNNs by co-augmenting topology and attribute
• Improves graph representation via contrastive learning

Innovation

• Dual-channel GNN for topology-attribute co-augmentation
• Contrastive learning with prototype alignment
• Bipartite graph projection for structure refinement