Dual Boost-Driven Graph-Level Clustering Network

2025-04-08

AI Summary

Graph-level clustering faces two core challenges: structural noise interference and representation distortion caused by cumulative noise during feature propagation. To address these, we propose a dual-driven collaborative optimization framework featuring a novel “representation learning–interference identification” joint boosting mechanism. First, we design a learnable global feature transformation matrix coupled with a similarity-guided adaptive interference suppression strategy to mitigate noise aggregation during pooling. Second, we introduce a multi-view similarity modeling module integrated with weighted graph pooling to enhance clustering discriminability. Implemented as an end-to-end trainable Graph Neural Network (GNN), our method achieves state-of-the-art performance across six benchmark datasets, consistently outperforming existing approaches in both clustering accuracy and robustness against structural noise.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Computer Vision: Representation Learning for Vision
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Graph-level clustering remains a pivotal yet formidable challenge in graph learning. Recently, the integration of deep learning with representation learning has demonstrated notable advancements, yielding performance enhancements to a certain degree. However, existing methods suffer from at least one of the following issues: 1. the original graph structure has noise, and 2. during feature propagation and pooling processes, noise is gradually aggregated into the graph-level embeddings through information propagation. Consequently, these two limitations mask clustering-friendly information, leading to suboptimal graph-level clustering performance. To this end, we propose a novel Dual Boost-Driven Graph-Level Clustering Network (DBGCN) to alternately promote graph-level clustering and filtering out interference information in a unified framework. Specifically, in the pooling step, we evaluate the contribution of features at the global and optimize them using a learnable transformation matrix to obtain high-quality graph-level representation, such that the model's reasoning capability can be improved. Moreover, to enable reliable graph-level clustering, we first identify and suppress information detrimental to clustering by evaluating similarities between graph-level representations, providing more accurate guidance for multi-view fusion. Extensive experiments demonstrated that DBGCN outperforms the state-of-the-art graph-level clustering methods on six benchmark datasets.

Problem

• Address noise in original graph structure
• Reduce noise aggregation in graph embeddings
• Enhance clustering-friendly information extraction

Innovation

• Dual boost-driven clustering and noise filtering
• Learnable transformation matrix for optimized representation
• Similarity-based suppression of detrimental clustering information