Exploiting Transitivity for Entity Matching

Jurian Baas, Mehdi Dastani, Ad Feelders

Keywords: Knowledge Graphs, Cluster Editing, Entity Matching

TL;DR: We propose a methodology for entity matching that starts with a given similarity measure, generates a set of entity pairs, and applies cluster editing to enforce transitivity.

Abstract: The goal of entity matching in knowledge graphs is to identify sets of entities that refer to the same real-world object. Methods for entity matching in knowledge graphs, however, produce a collection of pairs of entities claimed to be duplicates. This collection may fail to satisfy transitivity, and hence may fail to represent a valid solution. We show that an ad-hoc enforcement of transitivity on the set of identified entity pairs may decrease precision dramatically. We therefore propose a methodology that starts with a given similarity measure, generates a set of entity pairs, and applies cluster editing to enforce transitivity, leading to overall improved performance.