Tri-Learn Graph Fusion Network for Attributed Graph Clustering

2025-07-17

AI Summary

To address over-smoothing and over-compression issues in Graph Convolutional Networks (GCNs) for attributed graph clustering, this paper proposes a novel tri-channel collaborative learning framework integrating GCNs, autoencoders, and graph Transformers. The framework jointly models node attributes and topological structure via a designed mutual-learning mechanism and a multi-channel feature fusion module, thereby enhancing both global consistency and local discriminability. Unlike monolithic architectures, it enables deep complementarity and joint optimization across the three representation learning paradigms. Extensive experiments on ACM, Reuters, and USPS datasets demonstrate absolute improvements of 0.87%, 14.14%, and 7.58% in clustering accuracy, respectively, outperforming state-of-the-art methods. These results validate the framework’s effectiveness and generalizability in real-world applications such as news categorization.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Computer Vision: Representation Learning for Vision
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

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

Abstract

In recent years, models based on Graph Convolutional Networks (GCN) have made significant strides in the field of graph data analysis. However, challenges such as over-smoothing and over-compression remain when handling large-scale and complex graph datasets, leading to a decline in clustering quality. Although the Graph Transformer architecture has mitigated some of these issues, its performance is still limited when processing heterogeneous graph data. To address these challenges, this study proposes a novel deep clustering framework that comprising GCN, Autoencoder (AE), and Graph Transformer, termed the Tri-Learn Graph Fusion Network (Tri-GFN). This framework enhances the differentiation and consistency of global and local information through a unique tri-learning mechanism and feature fusion enhancement strategy. The framework integrates GCN, AE, and Graph Transformer modules. These components are meticulously fused by a triple-channel enhancement module, which maximizes the use of both node attributes and topological structures, ensuring robust clustering representation. The tri-learning mechanism allows mutual learning among these modules, while the feature fusion strategy enables the model to capture complex relationships, yielding highly discriminative representations for graph clustering. It surpasses many state-of-the-art methods, achieving an accuracy improvement of approximately 0.87% on the ACM dataset, 14.14 % on the Reuters dataset, and 7.58 % on the USPS dataset. Due to its outstanding performance on the Reuters dataset, Tri-GFN can be applied to automatic news classification, topic retrieval, and related fields.

Problem

• Overcoming over-smoothing and over-compression in GCN-based graph clustering
• Improving performance on heterogeneous graph data with Graph Transformer
• Enhancing clustering accuracy via tri-learning and feature fusion mechanisms

Innovation

• Combines GCN, AE, and Graph Transformer modules
• Uses tri-learning mechanism for mutual learning
• Enhances feature fusion for discriminative representations