Domain Representation for Knowledge Graph Embedding

TL;DR

This paper introduces domain representations in knowledge graph embeddings to capture hierarchical domain similarities, significantly improving link prediction accuracy.

Contribution

It proposes a novel method to learn domain embeddings over existing models, capturing hierarchical knowledge and enhancing link prediction performance.

Findings

Domain embeddings improve link prediction accuracy.

Hierarchical domain knowledge enhances embedding quality.

Significant performance gain over state-of-the-art models.

Abstract

Embedding entities and relations into a continuous multi-dimensional vector space have become the dominant method for knowledge graph embedding in representation learning. However, most existing models ignore to represent hierarchical knowledge, such as the similarities and dissimilarities of entities in one domain. We proposed to learn a Domain Representations over existing knowledge graph embedding models, such that entities that have similar attributes are organized into the same domain. Such hierarchical knowledge of domains can give further evidence in link prediction. Experimental results show that domain embeddings give a significant improvement over the most recent state-of-art baseline knowledge graph embedding models.