Self-Supervised Contrastive Graph Clustering Network via Structural Information Fusion

TL;DR

This paper introduces CGCN, a deep graph clustering method that integrates contrastive learning and structural information to improve the reliability of pre-trained clustering distributions and enhance overall clustering performance.

Contribution

The paper proposes a novel contrastive graph clustering network that incorporates deep structural signals and adaptive information aggregation, addressing limitations of existing pre-training based methods.

Findings

Enhanced clustering reliability demonstrated on real-world datasets.

Significant performance improvements over traditional graph clustering methods.

Effective utilization of contrastive signals for better module interoperability.

Abstract

Graph clustering, a classical task in graph learning, involves partitioning the nodes of a graph into distinct clusters. This task has applications in various real-world scenarios, such as anomaly detection, social network analysis, and community discovery. Current graph clustering methods commonly rely on module pre-training to obtain a reliable prior distribution for the model, which is then used as the optimization objective. However, these methods often overlook deeper supervised signals, leading to sub-optimal reliability of the prior distribution. To address this issue, we propose a novel deep graph clustering method called CGCN. Our approach introduces contrastive signals and deep structural information into the pre-training process. Specifically, CGCN utilizes a contrastive learning mechanism to foster information interoperability among multiple modules and allows the model to adaptively adjust the degree of information aggregation for different order structures. Our CGCN method has been experimentally validated on multiple real-world graph datasets, showcasing its ability to boost the dependability of prior clustering distributions acquired through pre-training. As a result, we observed notable enhancements in the performance of the model.