Robust Contrastive Graph Clustering with Adaptive Local-Global Integration

2026-05-27

AI Summary

Existing graph clustering methods struggle to simultaneously capture high-order local structures and global semantics, leading to limited performance on graphs with fragmented topology and ambiguous cluster boundaries. This work proposes a contrastive graph clustering framework that innovatively integrates multi-scale GNN-derived local features with dynamically evolving global semantic prototypes. The approach employs an attention mechanism to adaptively weight topological information from multiple propagation depths and leverages cluster centroids to guide semantic aggregation. Furthermore, it introduces a dual-view contrastive learning strategy that jointly optimizes instance-level discriminability and structure-aware objectives. Evaluated on eight real-world graph datasets, the method significantly outperforms state-of-the-art approaches, effectively enhancing representation robustness and inter-cluster separability.

Abstract

Graph clustering is essential in graph analysis for revealing structural patterns and node communities. Despite recent advances in self-supervised contrastive learning that have improved clustering via structural and attribute signals, existing methods still struggle to flexibly capture high-order local structures and often overlook global semantics in complex graphs. These limitations lead to suboptimal node representations, especially in real-world graphs with fragmented structures and ambiguous cluster boundaries. To address these limitations, a contrastive graph clustering framework is proposed to jointly integrate multi-scale local structures with global semantics via attention mechanisms. At the local level, GNN-based topological signals extracted from multiple propagation depths are adaptively fused through attention-based weighting to capture multi-scale neighborhood features. At the global level, semantic prototypes derived from dynamically evolving cluster centers are adaptively aggregated through attention to guide node representations and enhance inter-cluster separability. The model is trained under a dual-view contrastive learning paradigm with a hybrid objective that combines instance-level and structure-aware losses to improve representation robustness and discrimination. Experiments on eight real-world graph datasets demonstrate that our method achieves competitive clustering performance. Code is available at https://github.com/vege12138/w2.

Problem

• graph clustering
• local-global integration
• high-order structures
• global semantics
• node representations

Innovation

• contrastive learning
• graph clustering
• local-global integration
• attention mechanism
• multi-scale representation