An Interpretable Knowledge Transfer Model for Knowledge Base Completion

TL;DR

The paper introduces ITransF, an interpretable embedding model with a sparse attention mechanism that enhances knowledge base completion by discovering hidden relation concepts and sharing statistical strength, leading to improved performance.

Contribution

ITransF is a novel, interpretable knowledge base completion model that uses sparse attention to discover hidden concepts and transfer knowledge across relations.

Findings

Improves mean rank and Hits@10 metrics on WN18 and FB15k datasets.

Provides interpretable relation-concept associations.

Outperforms baseline models without additional information.

Abstract

Knowledge bases are important resources for a variety of natural language processing tasks but suffer from incompleteness. We propose a novel embedding model, \emph{ITransF}, to perform knowledge base completion. Equipped with a sparse attention mechanism, ITransF discovers hidden concepts of relations and transfer statistical strength through the sharing of concepts. Moreover, the learned associations between relations and concepts, which are represented by sparse attention vectors, can be interpreted easily. We evaluate ITransF on two benchmark datasets---WN18 and FB15k for knowledge base completion and obtains improvements on both the mean rank and Hits@10 metrics, over all baselines that do not use additional information.