Unifying Post-hoc Explanations of Knowledge Graph Completions

2025-07-29

AI Summary

Post-hoc explainability in knowledge graph completion (KGC) lacks a formal evaluation framework, hindering method reproducibility and cross-study comparison. Method: This paper introduces the first multi-objective optimization formulation for post-hoc KGC explanation, proposing a unified, generic explanation model that jointly optimizes explanation effectiveness (e.g., MRR, Hits@k), conciseness, and relevance to user queries. A standardized evaluation protocol is established to enable quantitative, comparable, and reproducible assessment of explanation quality. Contribution/Results: Extensive experiments demonstrate the framework’s broad applicability across diverse KGC models and benchmark datasets. It significantly enhances the theoretical rigor and empirical comparability of post-hoc explanation methods, establishing a new benchmark for explainable KGC. The framework bridges critical gaps between interpretability research and practical deployment, enabling systematic evaluation and advancement of transparent KGC systems.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Machine Learning: Learning Preferences or Rankings

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

Post-hoc explainability for Knowledge Graph Completion (KGC) lacks formalization and consistent evaluations, hindering reproducibility and cross-study comparisons. This paper argues for a unified approach to post-hoc explainability in KGC. First, we propose a general framework to characterize post-hoc explanations via multi-objective optimization, balancing their effectiveness and conciseness. This unifies existing post-hoc explainability algorithms in KGC and the explanations they produce. Next, we suggest and empirically support improved evaluation protocols using popular metrics like Mean Reciprocal Rank and Hits@$k$. Finally, we stress the importance of interpretability as the ability of explanations to address queries meaningful to end-users. By unifying methods and refining evaluation standards, this work aims to make research in KGC explainability more reproducible and impactful.

Problem

• Lack of formalization in post-hoc KGC explainability evaluations
• Need for unified framework balancing explanation effectiveness and conciseness
• Improving interpretability by addressing user-relevant queries in KGC

Innovation

• Multi-objective optimization for explanation framework
• Improved evaluation protocols with popular metrics
• Focus on interpretability for end-user queries