Compactness and Consistency: A Conjoint Framework for Deep Graph Clustering

2026-10-08

AI Summary

This study addresses the limitations of Graph Neural Networks (GNNs) in capturing global dependencies through local message passing, as well as the lack of compactness and robustness in node representations caused by redundant noise in graph data. To this end, we propose the CoCo framework, which employs graph convolutional filters to extract dual-perspective features from both local and global views. These features are then encoded into compact embeddings via low-rank matrix factorization, effectively denoising the data and revealing the intrinsic graph structure. Furthermore, a consistency regularization strategy is introduced to facilitate cross-view knowledge transfer and enhance semantic expressiveness. Extensive experiments demonstrate that CoCo significantly outperforms existing state-of-the-art models across multiple benchmark datasets.

Abstract

Graph clustering is a fundamental task in data analysis, aiming at grouping nodes with similar characteristics in the graph into clusters. This problem has been widely explored using graph neural networks (GNNs) due to their ability to leverage node attributes and graph topology for effective cluster assignments. However, representations learned through GNNs typically struggle to capture global relationships between nodes via local message-passing mechanisms. Moreover, the redundancy and noise inherently present in graph data may easily result in node representations lacking compactness and robustness. To address these issues, we propose a conjoint framework CoCo, which captures compactness and consistency in the learned node representations for deep graph clustering. Technically, our CoCo leverages graph convolutional filters to learn robust node representations from both local and global views, and then encodes them into low-rank compact embeddings, thus effectively removing the redundancy and noise as well as uncovering the intrinsic underlying structure. To further enrich the node semantics, we develop a consistency learning strategy based on compact embeddings to facilitate knowledge transfer from the two perspectives. Our experimental results indicate that our CoCo outperforms state-of-the-art counterparts on various datasets.

Problem

• Deep Graph Clustering
• Graph Neural Networks
• Node Representation
• Compactness
• Consistency

Innovation

• Deep Graph Clustering
• Graph Convolutional Filters
• Low-rank Compact Embeddings
• Consistency Learning
• Conjoint Framework