TranS: Transition-based Knowledge Graph Embedding with Synthetic Relation Representation

TL;DR

This paper introduces TranS, a novel transition-based knowledge graph embedding method that replaces single relation vectors with synthetic representations, effectively handling complex relation scenarios and achieving state-of-the-art results.

Contribution

The paper proposes a new transition-based KGE model, TranS, which uses synthetic relation representations to improve performance on complex relation types.

Findings

Achieves state-of-the-art results on ogbl-wikikg2 dataset.

Effectively handles complex relation scenarios with synthetic relation representations.

Outperforms previous models in knowledge graph embedding tasks.

Abstract

Knowledge graph embedding (KGE) aims to learn continuous vectors of relations and entities in knowledge graph. Recently, transition-based KGE methods have achieved promising performance, where the single relation vector learns to translate head entity to tail entity. However, this scoring pattern is not suitable for complex scenarios where the same entity pair has different relations. Previous models usually focus on the improvement of entity representation for 1-to-N, N-to-1 and N-to-N relations, but ignore the single relation vector. In this paper, we propose a novel transition-based method, TranS, for knowledge graph embedding. The single relation vector in traditional scoring patterns is replaced with synthetic relation representation, which can solve these issues effectively and efficiently. Experiments on a large knowledge graph dataset, ogbl-wikikg2, show that our model achieves state-of-the-art results.