Rel-HNN: Split Parallel Hypergraph Neural Network for Learning on Relational Databases

2025-07-16

AI Summary

Existing graph neural networks (GNNs) for relational databases suffer from insufficient fine-grained intra-tuple semantic modeling, as they treat entire tuples as atomic nodes and neglect attribute-level interactions. To address this, we propose HGNN-Rel, a hypergraph neural network that explicitly captures intra-tuple structural semantics by modeling attribute-value pairs as nodes and tuples as hyperedges. HGNN-Rel introduces a multi-level representation learning framework that jointly encodes attribute-, tuple-, and table-level information in a unified manner. Furthermore, we design a split-and-parallel training algorithm to enable efficient multi-GPU scalability. Extensive experiments on multiple real-world and benchmark relational datasets demonstrate that HGNN-Rel significantly outperforms state-of-the-art methods on both classification and regression tasks. It achieves up to 3.18× faster training speed and 2.94× higher hypergraph learning efficiency, striking an effective balance between modeling accuracy and computational scalability.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Relational Probabilistic Models
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Relational databases (RDBs) are ubiquitous in enterprise and real-world applications. Flattening the database poses challenges for deep learning models that rely on fixed-size input representations to capture relational semantics from the structured nature of relational data. Graph neural networks (GNNs) have been proposed to address this, but they often oversimplify relational structures by modeling all the tuples as monolithic nodes and ignoring intra-tuple associations. In this work, we propose a novel hypergraph-based framework, that we call rel-HNN, which models each unique attribute-value pair as a node and each tuple as a hyperedge, enabling the capture of fine-grained intra-tuple relationships. Our approach learns explicit multi-level representations across attribute-value, tuple, and table levels. To address the scalability challenges posed by large RDBs, we further introduce a split-parallel training algorithm that leverages multi-GPU execution for efficient hypergraph learning. Extensive experiments on real-world and benchmark datasets demonstrate that rel-HNN significantly outperforms existing methods in both classification and regression tasks. Moreover, our split-parallel training achieves substantial speedups -- up to 3.18x for learning on relational data and up to 2.94x for hypergraph learning -- compared to conventional single-GPU execution.

Problem

• Capturing relational semantics from structured data efficiently
• Modeling fine-grained intra-tuple relationships in databases
• Scaling hypergraph learning for large relational databases

Innovation

• Hypergraph-based framework for fine-grained relational learning
• Multi-level representations across attribute, tuple, table
• Split-parallel multi-GPU training for scalable hypergraph learning