StruProKGR: A Structural and Probabilistic Framework for Sparse Knowledge Graph Reasoning

2025-12-14

AI Summary

Sparse knowledge graph (KG) reasoning faces significant challenges, including severe fact incompleteness and difficulty in capturing complex relational patterns. To address these issues, we propose a novel path-based reasoning framework. First, we design a distance-guided, structure-aware path sampling mechanism to alleviate path scarcity induced by KG sparsity. Second, we introduce a probabilistic structural path aggregation model that jointly leverages graph-distance metrics and path-confidence weighting, overcoming the limitations of conventional independent-path modeling. Our approach achieves strong interpretability and computational efficiency simultaneously. Extensive experiments on five sparse KG benchmarks demonstrate that it significantly outperforms existing path-based methods: average inference accuracy improves substantially, and inference latency decreases by over 50%. To the best of our knowledge, this is the first method to achieve an organic integration of high accuracy, low computational overhead, and inherent interpretability in sparse KG reasoning.

Technology Category

• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Reasoning under Uncertainty: Relational Probabilistic Models
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Sparse Knowledge Graphs (KGs) are commonly encountered in real-world applications, where knowledge is often incomplete or limited. Sparse KG reasoning, the task of inferring missing knowledge over sparse KGs, is inherently challenging due to the scarcity of knowledge and the difficulty of capturing relational patterns in sparse scenarios. Among all sparse KG reasoning methods, path-based ones have attracted plenty of attention due to their interpretability. Existing path-based methods typically rely on computationally intensive random walks to collect paths, producing paths of variable quality. Additionally, these methods fail to leverage the structured nature of graphs by treating paths independently. To address these shortcomings, we propose a Structural and Probabilistic framework named StruProKGR, tailored for efficient and interpretable reasoning on sparse KGs. StruProKGR utilizes a distance-guided path collection mechanism to significantly reduce computational costs while exploring more relevant paths. It further enhances the reasoning process by incorporating structural information through probabilistic path aggregation, which prioritizes paths that reinforce each other. Extensive experiments on five sparse KG reasoning benchmarks reveal that StruProKGR surpasses existing path-based methods in both effectiveness and efficiency, providing an effective, efficient, and interpretable solution for sparse KG reasoning.

Problem

• Addresses inefficiency in path collection for sparse knowledge graphs
• Improves reasoning by incorporating structural information probabilistically
• Enhances interpretability and effectiveness in sparse KG inference

Innovation

• Distance-guided path collection reduces computational costs
• Probabilistic path aggregation leverages structural graph information
• Framework enhances sparse knowledge graph reasoning efficiency and interpretability