Learning Description Logic Ontologies. Five Approaches. Where Do They Stand?

TL;DR

This paper reviews five different approaches to automatically learning description logic ontologies, comparing their methods, benefits, and limitations to advance formal knowledge representation.

Contribution

It provides a comprehensive overview and comparison of five classical machine learning and data mining approaches for DL ontology learning.

Findings

Each approach has unique strengths and limitations.

Neural networks and association rule mining are promising for DL ontology learning.

No single approach is universally best; hybrid methods may be needed.

Abstract

The quest for acquiring a formal representation of the knowledge of a domain of interest has attracted researchers with various backgrounds into a diverse field called ontology learning. We highlight classical machine learning and data mining approaches that have been proposed for (semi-)automating the creation of description logic (DL) ontologies. These are based on association rule mining, formal concept analysis, inductive logic programming, computational learning theory, and neural networks. We provide an overview of each approach and how it has been adapted for dealing with DL ontologies. Finally, we discuss the benefits and limitations of each of them for learning DL ontologies.