header

Expressive Ontology Learning as Neural Machine Translation

26 Pages Posted: 29 Oct 2018 First Look: Accepted

See all articles by Giulio Petrucci

Giulio Petrucci

Fondazione Bruno Kessler

Marco Rospocher

Fondazione Bruno Kessler

Chiara Ghidini

Fondazione Bruno Kessler

Abstract

Automated ontology learning from unstructured textual sources has been proposed in literature as a way to support the difficult and time-consuming task of knowledge modeling for semantic applications. In this paper we propose a system, based on a neural network in the encoder-decoder configuration, to translate natural language definitions into Description Logics formulae through syntactic transformation. The model has been evaluated to asses its capacity to generalize over different syntactic structures, tolerate unknown words, and improve its performance by enriching the training set with new annotated examples. The results obtained in our evaluation show how approaching the ontology learning problem as a neural machine translation task can be a valid way to tackle long term expressive ontology learning challenges such as language variability, domain independence, and high engineering costs.

Keywords: ontology learning, neural networks, natural language processing

Suggested Citation

Petrucci, Giulio and Rospocher, Marco and Ghidini, Chiara, Expressive Ontology Learning as Neural Machine Translation (October 29, 2018). Journal of Web Semantics First Look . Available at SSRN: https://ssrn.com/abstract=3274562 or http://dx.doi.org/10.2139/ssrn.3274562

Giulio Petrucci (Contact Author)

Fondazione Bruno Kessler

Via Sommarive 18
Povo
Trento, 38123
Italy

Marco Rospocher

Fondazione Bruno Kessler

Via Sommarive 18
Povo
Trento, 38123
Italy

Chiara Ghidini

Fondazione Bruno Kessler

Via Sommarive 18
Povo
Trento, 38123
Italy