Modeling Relation Paths for Representation Learning of Knowledge Bases

TL;DR

This paper introduces a path-based representation learning model for knowledge bases that considers multi-step relation paths, improving inference and completion tasks by measuring path reliability and semantic composition.

Contribution

It proposes a novel model that incorporates relation paths as translations, with a path-constraint resource allocation algorithm and semantic composition for improved knowledge base embedding.

Findings

Significant improvements in knowledge base completion

Enhanced relation extraction from text

Effective measurement of relation path reliability

Abstract

Representation learning of knowledge bases (KBs) aims to embed both entities and relations into a low-dimensional space. Most existing methods only consider direct relations in representation learning. We argue that multiple-step relation paths also contain rich inference patterns between entities, and propose a path-based representation learning model. This model considers relation paths as translations between entities for representation learning, and addresses two key challenges: (1) Since not all relation paths are reliable, we design a path-constraint resource allocation algorithm to measure the reliability of relation paths. (2) We represent relation paths via semantic composition of relation embeddings. Experimental results on real-world datasets show that, as compared with baselines, our model achieves significant and consistent improvements on knowledge base completion and relation extraction from text.