Improving rule mining via embedding-based link prediction

TL;DR

This paper introduces a novel method that enhances rule mining in knowledge graphs by leveraging pre-trained embeddings to improve the discovery of valuable rules, combining interpretability with generalization.

Contribution

The authors propose enriching knowledge graphs with pre-trained embeddings before rule mining, offering a new hybrid approach that addresses convergence issues of unified models.

Findings

Discoveries of new valuable rules on enriched graphs

Improved rule mining performance on benchmark datasets

Open source implementation and pretrained models provided

Abstract

Rule mining on knowledge graphs allows for explainable link prediction. Contrarily, embedding-based methods for link prediction are well known for their generalization capabilities, but their predictions are not interpretable. Several approaches combining the two families have been proposed in recent years. The majority of the resulting hybrid approaches are usually trained within a unified learning framework, which often leads to convergence issues due to the complexity of the learning task. In this work, we propose a new way to combine the two families of approaches. Specifically, we enrich a given knowledge graph by means of its pre-trained entity and relation embeddings before applying rule mining systems on the enriched knowledge graph. To validate our approach, we conduct extensive experiments on seven benchmark datasets. An analysis of the results generated by our approach suggests that we discover new valuable rules on the enriched graphs. We provide an open source implementation of our approach as well as pretrained models and datasets at https://github.com/Jean-KOUAGOU/EnhancedRuleLearning