RAW-Explainer: Post-hoc Explanations of Graph Neural Networks on Knowledge Graphs

2025-06-14

AI Summary

Explainability of graph neural networks (GNNs) for knowledge graph (KG) link prediction faces key challenges in heterogeneous environments: difficulty in generating faithful explanations, poor subgraph connectivity, and severe distribution shift between local explanations and the global KG. To address these, we propose the first neural-parameterized subgraph explanation framework specifically designed for link prediction. Our method identifies factually grounded explanation patterns via heterogeneous semantic-aware random walks, yielding connected and concise subgraph explanations. We further introduce a robust subgraph evaluator that adaptively calibrates distributional discrepancies between explanatory subgraphs and the full KG. Evaluated on multiple real-world KG benchmarks, our approach achieves state-of-the-art trade-offs: subgraph connectivity improves by 37%, inference speed reaches 5.2× that of baseline methods, and performance significantly surpasses existing node- or graph-level explainability approaches.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for 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 learning to rank, online learning, and counterfactual learning for ranking

Abstract

Graph neural networks have demonstrated state-of-the-art performance on knowledge graph tasks such as link prediction. However, interpreting GNN predictions remains a challenging open problem. While many GNN explainability methods have been proposed for node or graph-level tasks, approaches for generating explanations for link predictions in heterogeneous settings are limited. In this paper, we propose RAW-Explainer, a novel framework designed to generate connected, concise, and thus interpretable subgraph explanations for link prediction. Our method leverages the heterogeneous information in knowledge graphs to identify connected subgraphs that serve as patterns of factual explanation via a random walk objective. Unlike existing methods tailored to knowledge graphs, our approach employs a neural network to parameterize the explanation generation process, which significantly speeds up the production of collective explanations. Furthermore, RAW-Explainer is designed to overcome the distribution shift issue when evaluating the quality of an explanatory subgraph which is orders of magnitude smaller than the full graph, by proposing a robust evaluator that generalizes to the subgraph distribution. Extensive quantitative results on real-world knowledge graph datasets demonstrate that our approach strikes a balance between explanation quality and computational efficiency.

Problem

• Interpreting GNN predictions for knowledge graphs
• Generating explanations for link predictions in heterogeneous settings
• Overcoming distribution shift in evaluating explanatory subgraphs

Innovation

• Random walk objective for connected subgraph explanations
• Neural network parameterizes explanation generation
• Robust evaluator addresses subgraph distribution shift