Abstract:
The hypergraph neural network (HGNN) is an emerging powerful tool for modeling and learning complex, high-order correlations among entities upon hypergraph structures. Wh...Show MoreMetadata
Abstract:
The hypergraph neural network (HGNN) is an emerging powerful tool for modeling and learning complex, high-order correlations among entities upon hypergraph structures. While existing HGNN-based approaches excel in modeling high-order correlations among data using hyperedges, they often have difficulties in distinguishing diverse semantics ( e.g., bioactivities between drug and target in biological networks) of different correlations, making it challenging to learn accurate final representations. The underlying reason is that the specific semantic information of each hyperedge cannot be captured and distinguished during the modeling and learning process. To address this, we propose a mode HGNN (MHGNN) framework that extends the vanilla hypergraph structure by endowing hyperedges with mode information for encapsulating their semantics and then performs mode-aware high-order message passing upon mode hypergraph for achieving comprehensive node representations. Extensive evaluations on four real-world datasets under two representative tasks have demonstrated the outstanding performance of MHGNN against the state of the arts.
Published in: IEEE Transactions on Neural Networks and Learning Systems ( Early Access )
Funding Agency:
Keywords assist with retrieval of results and provide a means to discovering other relevant content. Learn more.
- IEEE Keywords
- Index Terms
- Semantic ,
- Learning Process ,
- Real-world Datasets ,
- Complex Learning ,
- Node Representations ,
- Message Passing ,
- Types Of Interactions ,
- Global Distribution ,
- Types Of Modes ,
- Heterogeneous Network ,
- Mode Of Distribution ,
- Graph Convolutional Network ,
- Node Features ,
- Graph Neural Networks ,
- Self-supervised Learning ,
- Embedding Methods ,
- Graph Convolution ,
- Graph-based Methods ,
- Link Prediction ,
- Node Embeddings ,
- Graph Attention Network ,
- Global Structural Information ,
- Locally Compact ,
- Mode Spacing ,
- Heterogeneous Graph ,
- Recommendation Task ,
- Different Types Of Modes ,
- Network Embedding ,
- Negative Samples ,
- Random Walk
- Author Keywords
Keywords assist with retrieval of results and provide a means to discovering other relevant content. Learn more.
- IEEE Keywords
- Index Terms
- Semantic ,
- Learning Process ,
- Real-world Datasets ,
- Complex Learning ,
- Node Representations ,
- Message Passing ,
- Types Of Interactions ,
- Global Distribution ,
- Types Of Modes ,
- Heterogeneous Network ,
- Mode Of Distribution ,
- Graph Convolutional Network ,
- Node Features ,
- Graph Neural Networks ,
- Self-supervised Learning ,
- Embedding Methods ,
- Graph Convolution ,
- Graph-based Methods ,
- Link Prediction ,
- Node Embeddings ,
- Graph Attention Network ,
- Global Structural Information ,
- Locally Compact ,
- Mode Spacing ,
- Heterogeneous Graph ,
- Recommendation Task ,
- Different Types Of Modes ,
- Network Embedding ,
- Negative Samples ,
- Random Walk
- Author Keywords