Look Globally and Reason: Two-stage Path Reasoning over Sparse Knowledge Graphs

TL;DR

This paper introduces LoGRe, a two-stage path reasoning model that leverages global analysis of training data to improve reasoning over sparse knowledge graphs, enhancing explainability and performance.

Contribution

LoGRe is a novel two-stage reasoning approach that constructs relation-path schemas from global data analysis, addressing sparsity without external models.

Findings

Outperforms existing models on five benchmark datasets

Effectively alleviates sparsity issues in knowledge graphs

Provides explainable reasoning paths

Abstract

Sparse Knowledge Graphs (KGs), frequently encountered in real-world applications, contain fewer facts in the form of (head entity, relation, tail entity) compared to more populated KGs. The sparse KG completion task, which reasons answers for given queries in the form of (head entity, relation, ?) for sparse KGs, is particularly challenging due to the necessity of reasoning missing facts based on limited facts. Path-based models, known for excellent explainability, are often employed for this task. However, existing path-based models typically rely on external models to fill in missing facts and subsequently perform path reasoning. This approach introduces unexplainable factors or necessitates meticulous rule design. In light of this, this paper proposes an alternative approach by looking inward instead of seeking external assistance. We introduce a two-stage path reasoning model called LoGRe (Look Globally and Reason) over sparse KGs. LoGRe constructs a relation-path reasoning schema by globally analyzing the training data to alleviate the sparseness problem. Based on this schema, LoGRe then aggregates paths to reason out answers. Experimental results on five benchmark sparse KG datasets demonstrate the effectiveness of the proposed LoGRe model.