Revisiting Dynamic Graph Clustering via Matrix Factorization

2025-02-10

AI Summary

Dynamic graph clustering aims to model the temporal evolution of node cluster structures, yet existing matrix factorization approaches suffer from poor scalability, weak noise robustness, and low computational efficiency. To address these challenges, we propose DyCluster—a scalable and robust dynamic graph clustering framework. Its key contributions are: (1) temporal-separation matrix factorization, which decouples static structural patterns from dynamic evolutionary trends; (2) dual-clustering regularization, jointly enforcing consistency across both node-level and time-level cluster assignments; and (3) selective embedding update, wherein only nodes exhibiting significant structural changes are re-optimized to reduce computational overhead. Extensive experiments on six synthetic and five real-world dynamic graph datasets demonstrate that DyCluster consistently outperforms state-of-the-art baselines in clustering accuracy, robustness, and efficiency. The implementation is publicly available.

Technology Category

• Machine Learning: Clustering
• Data Mining & Knowledge Management: Scalability, Parallel & Distributed Systems
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Dynamic graph clustering aims to detect and track time-varying clusters in dynamic graphs, revealing the evolutionary mechanisms of complex real-world dynamic systems. Matrix factorization-based methods are promising approaches for this task; however, these methods often struggle with scalability and can be time-consuming when applied to large-scale dynamic graphs. Moreover, they tend to lack robustness and are vulnerable to real-world noisy data. To address these issues, we make three key contributions. First, to improve scalability, we propose temporal separated matrix factorization, where a single matrix is divided into multiple smaller matrices for independent factorization, resulting in faster computation. Second, to improve robustness, we introduce bi-clustering regularization, which jointly optimizes graph embedding and clustering, thereby filtering out noisy features from the graph embeddings. Third, to further enhance effectiveness and efficiency, we propose selective embedding updating, where we update only the embeddings of dynamic nodes while the embeddings of static nodes are fixed among different timestamps. Experimental results on six synthetic and five real-world benchmarks demonstrate the scalability, robustness and effectiveness of our proposed method. Source code is available at https://github.com/Clearloveyuan/DyG-MF.

Problem

• Dynamic graph clustering challenges
• Scalability and robustness issues
• Noise vulnerability in real data

Innovation

• Temporal separated matrix factorization
• Bi-clustering regularization
• Selective embedding updating