MetaHGNIE: Meta-Path Induced Hypergraph Contrastive Learning in Heterogeneous Knowledge Graphs

2025-12-13

AI Summary

Existing node importance estimation (NIE) methods for heterogeneous knowledge graphs struggle to capture high-order dependencies and fail to jointly model structural and semantic signals. Method: We propose a meta-path-guided structure–semantics disentangled alignment framework: (i) explicitly modeling meta-path sequences as typed hyperedges to construct a semantics-aware high-order hypergraph; (ii) designing a sparse block-wise hypergraph Transformer for efficient local–global structural aggregation; and (iii) integrating multimodal contrastive learning with auxiliary supervision to jointly optimize cross-modal representations. Contribution/Results: Our approach achieves significant improvements over state-of-the-art methods across multiple NIE benchmarks, demonstrating the effectiveness of high-order interaction modeling and synergistic structure–semantics alignment.

Technology Category

• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Machine Learning: Graph-based Machine Learning
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Bridging structured and unstructured data

Abstract

Node importance estimation (NIE) in heterogeneous knowledge graphs is a critical yet challenging task, essential for applications such as recommendation, knowledge reasoning, and question answering. Existing methods often rely on pairwise connections, neglecting high-order dependencies among multiple entities and relations, and they treat structural and semantic signals independently, hindering effective cross-modal integration. To address these challenges, we propose MetaHGNIE, a meta-path induced hypergraph contrastive learning framework for disentangling and aligning structural and semantic information. MetaHGNIE constructs a higher-order knowledge graph via meta-path sequences, where typed hyperedges capture multi-entity relational contexts. Structural dependencies are aggregated with local attention, while semantic representations are encoded through a hypergraph transformer equipped with sparse chunking to reduce redundancy. Finally, a multimodal fusion module integrates structural and semantic embeddings under contrastive learning with auxiliary supervision, ensuring robust cross-modal alignment. Extensive experiments on benchmark NIE datasets demonstrate that MetaHGNIE consistently outperforms state-of-the-art baselines. These results highlight the effectiveness of explicitly modeling higher-order interactions and cross-modal alignment in heterogeneous knowledge graphs. Our code is available at https://github.com/SEU-WENJIA/DualHNIE

Problem

• Estimating node importance in heterogeneous knowledge graphs
• Addressing neglect of high-order dependencies among entities
• Integrating structural and semantic information effectively

Innovation

• Meta-path induced hypergraph construction for higher-order dependencies
• Hypergraph transformer with sparse chunking for semantic encoding
• Contrastive learning with multimodal fusion for cross-modal alignment