Pushing the limits of OWL 2 reasoners in ontology alignment repair problems

Abstract

Ontologies play a key role in the development of the Semantic Web and are being used in many diverse application domains such as biomedicine and e-commerce. An application domain may have been modeled according to different points of view and purposes. This situation usually leads to the development of different ontologies that intuitively overlap, but that use different naming and modeling conventions. The problem of (semi-)automatically integrating independently developed ontologies through mappings, is usually referred to as the ontology matching problem. Ontology matching systems, however, rely on lexical and structural heuristics, and the integration of the input ontologies and the mappings may lead to many undesired logical consequences, which could sensibly diminish their usefulness. The present paper, on the one hand aims at verifying the hypothesis that classification of large ontologies via mappings still poses a challenge to OWL 2 reasoners. On the other it also explores the applicability of OWL 2 reasoning for the repair of unintended entailments (namely, unsatisfiable concepts or violations of the conservativity principle). In this paper we provide an update on the feasibility of using OWL 2 reasoners to repair the integration of ontologies via mappings, providing a more accurate evaluation of the feasibility of extracting all the justifications. Additionally, the current evaluation also encompasses the analysis of the use of OWL 2 reasoners for solving the violations of the so-called conservativity principle.