Hyperbolic Heterogeneous Graph Attention Networks
Open Access
- 13 May 2024
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 561-564
Abstract
Most previous heterogeneous graph embedding models represent elements in a heterogeneous graph as vector representations in a low-dimensional Euclidean space. However, because heterogeneous graphs inherently possess complex structures, such as hierarchical or power-law structures, distortions can occur when representing them in Euclidean space. To overcome this limitation, we propose Hyperbolic Heterogeneous Graph Attention Networks (HHGAT) that learn vector representations in hyperbolic spaces with metapath instances. We conducted experiments on three real-world heterogeneous graph datasets, demonstrating that HHGAT outperforms state-of-the-art heterogeneous graph embedding models in node classification and clustering tasks. This superior performance is attributed to HHGAT's ability to capture the complex structure of heterogeneous graphs effectively.Keywords
All Related Versions
Funding Information
- Institute of Information & Communications Technology Planning & Evaluation (RS-2022-00155857)
- National Research Foundation of Korea (RS-2023-00214065)
This publication has 10 references indexed in Scilit:
- Multi-Order Relations Hyperbolic Fusion for Heterogeneous Graphs2023
- MIGTNet: Metapath Instance-based Graph Transformation Network for heterogeneous graph embeddingFuture Generation Computer Systems, 2023
- Are we really making much progress?2021
- Embedding Heterogeneous Networks into Hyperbolic Space Without Meta-pathProceedings of the AAAI Conference on Artificial Intelligence, 2021
- GraphMSE: Efficient Meta-path Selection in Semantically Aligned Feature Space for Graph Neural NetworksProceedings of the AAAI Conference on Artificial Intelligence, 2021
- MAGNN: Metapath Aggregated Graph Neural Network for Heterogeneous Graph Embedding2020
- Heterogeneous Graph Transformer2020
- An Attention-Based Graph Neural Network for Heterogeneous Structural LearningProceedings of the AAAI Conference on Artificial Intelligence, 2020
- Hyperbolic Heterogeneous Information Network EmbeddingProceedings of the AAAI Conference on Artificial Intelligence, 2019
- Heterogeneous Graph Attention Network2019