Ontology Matching Through Absolute Orientation of Embedding Spaces

TL;DR

This paper introduces a novel ontology matching method using knowledge graph embeddings and absolute orientation to align embedding spaces, demonstrating effectiveness especially on similarly structured graphs and robustness to noise.

Contribution

It presents a new structure-based ontology matching approach leveraging absolute orientation of embeddings, with initial evaluation on synthetic and real datasets.

Findings

Works well on similarly structured graphs

Handles alignment noise better than size and structural differences

Preliminary results show promising effectiveness

Abstract

Ontology matching is a core task when creating interoperable and linked open datasets. In this paper, we explore a novel structure-based mapping approach which is based on knowledge graph embeddings: The ontologies to be matched are embedded, and an approach known as absolute orientation is used to align the two embedding spaces. Next to the approach, the paper presents a first, preliminary evaluation using synthetic and real-world datasets. We find in experiments with synthetic data, that the approach works very well on similarly structured graphs; it handles alignment noise better than size and structural differences in the ontologies.