Expressing OWL axioms by English sentences: dubious in theory, feasible in practice

With OWL (Web Ontology Language) established as a standard for encoding ontologies on the Semantic Web, interest has begun to focus on the task of verbalising OWL code in controlled English (or other natural language). Current approaches to this task assume that axioms in OWL can be mapped to sentences in English. We examine three potential problems with this approach (concerning logical sophistication, information structure, and size), and show that although these could in theory lead to insuperable difficulties, in practice they seldom arise, because ontology developers use OWL in ways that favour a transparent mapping. This result is evidenced by an analysis of patterns from a corpus of over 600,000 axioms in about 200 ontologies.

• Type: Journal
• Year: 2010
• Where: COLING
• Authors: Richard Power, Allan Third
• Keywords: COLING 2010, Computational Linguistics, Current Approaches, OWL Code, Web Ontology Language