NALDO: From natural language definitions to OWL expressions

Abstract

Domain ontologies are pivotal for Semantic Web applications. The richness of an ontology goes in hand with its usefulness and efficiency. Unfortunately, manually enriching an ontology is very time-consuming. In this paper, we propose to enrich an ontology automatically by obtaining logical expressions of concepts. We present NALDO, a novel approach that provides an OWL DL (Web Ontology Language Description Logics) expression of a concept from two inputs: (1) the natural language definition of the concept and (2) an ontology describing the domain of this concept. NALDO uses as much as possible entities provided by the domain ontology, however it can suggest, when needed, new entities. The expressiveness of expressions provided by NALDO covers value and cardinality restrictions, subsumption and equivalence. We evaluate our approach against the definitions and the corresponding ontologies of the BEAUFORD benchmark. Our results show that NALDO is able to perform the correct identification of formal entities with an F1-measure up to 0.79.