A Critical Examination of RESCAL for Completion of Knowledge Bases with Transitive Relations

TL;DR

This paper critically examines the RESCAL model for link prediction in knowledge graphs, revealing its inability to encode asymmetric transitive relations, which impacts its effectiveness in certain knowledge base properties.

Contribution

The paper provides a theoretical analysis of RESCAL, demonstrating its limitations in representing asymmetric transitive relations in knowledge bases.

Findings

RESCAL cannot encode asymmetric transitive relations

The analysis links knowledge base properties to model performance

Highlights limitations of RESCAL for certain relation types

Abstract

Link prediction in large knowledge graphs has received a lot of attention recently because of its importance for inferring missing relations and for completing and improving noisily extracted knowledge graphs. Over the years a number of machine learning researchers have presented various models for predicting the presence of missing relations in a knowledge base. Although all the previous methods are presented with empirical results that show high performance on select datasets, there is almost no previous work on understanding the connection between properties of a knowledge base and the performance of a model. In this paper we analyze the RESCAL method and prove that it can not encode asymmetric transitive relations in knowledge bases.