Multi-Relation Graph-Kernel Strengthen Network for Graph-Level Clustering

2025-04-02

AI Summary

To address the dual challenges of weak structural representation learning for unlabeled graph data clustering and the inability of conventional graph kernels to adaptively model multi-relational graphs, this paper proposes the first synergistic framework integrating multi-relational graph modeling with graph kernel techniques. Methodologically, we construct a multi-relational graph, design a relation-aware cross-view alignment mechanism and a progressive feature fusion strategy, and jointly incorporate subgraph/path kernel computation, relation-aware attention aggregation, multi-view contrastive distillation, and differentiable graph pooling. Extensive experiments on multiple benchmark datasets demonstrate that our approach significantly outperforms state-of-the-art methods, achieving substantial improvements in both graph-level embedding robustness and clustering accuracy. These results empirically validate the effectiveness of the novel “multi-relational modeling + kernel enhancement” paradigm.

Technology Category

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

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning 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: Bridging structured and unstructured data

Abstract

Graph-level clustering is a fundamental task of data mining, aiming at dividing unlabeled graphs into distinct groups. However, existing deep methods that are limited by pooling have difficulty extracting diverse and complex graph structure features, while traditional graph kernel methods rely on exhaustive substructure search, unable to adaptive handle multi-relational data. This limitation hampers producing robust and representative graph-level embeddings. To address this issue, we propose a novel Multi-Relation Graph-Kernel Strengthen Network for Graph-Level Clustering (MGSN), which integrates multi-relation modeling with graph kernel techniques to fully leverage their respective advantages. Specifically, MGSN constructs multi-relation graphs to capture diverse semantic relationships between nodes and graphs, which employ graph kernel methods to extract graph similarity features, enriching the representation space. Moreover, a relation-aware representation refinement strategy is designed, which adaptively aligns multi-relation information across views while enhancing graph-level features through a progressive fusion process. Extensive experiments on multiple benchmark datasets demonstrate the superiority of MGSN over state-of-the-art methods. The results highlight its ability to leverage multi-relation structures and graph kernel features, establishing a new paradigm for robust graph-level clustering.

Problem

• Improving graph-level clustering by integrating multi-relation modeling and graph kernels
• Overcoming limitations of deep methods and traditional graph kernel techniques
• Enhancing graph-level embeddings through adaptive multi-relation information fusion

Innovation

• Multi-relation modeling with graph kernels
• Relation-aware representation refinement strategy
• Progressive fusion for enhanced graph features