Title: Merging ontology by semantic enrichment and combining similarity measures

Authors: Messaouda Fareh; Omar Boussaid; Rachid Chalal; Melyara Mezzi; Khadija Nadji

Addresses: LRDSI Laboratory, Blida University, Soumaa, Algeria ' ERIC Laboratory, Lyon 2 University, Campus Porte des Alpes, France ' LMCS Laboratory, ESI, Oued-Smar (Algier), Algeria ' LRDSI Laboratory, Blida University, Soumaa, Algeria ' LRDSI Laboratory, Blida University, Soumaa, Algeria

Abstract: In this paper, we present a new approach to merge OWL ontologies by semantic enrichment of initial ontologies. This work is situated in the general context of stored information heterogeneity in a decisional system such as data, metadata and knowledge, for combination and reconciliation these forms of information by mediation. To add a semantic dimension to the merger, our approach based on semantic enrichment of initial ontologies, this is achieved by enriching initial ontologies by a set of metadata that annotate their concepts with synonyms and homonyms for each concept, via the use of WordNet, or semantic enrichment of an expert, then it generates a thesaurus for each local ontology to build the global thesaurus. Our method focuses on computing semantic similarity between concepts of ontologies, and based on a weighted combination of computing similarity methods, we use syntactic, lexical, structural and semantic techniques, for generating the correspondence matrix; from this latter we generate the merged ontology.

Keywords: merging ontologies; taxonomy; similarity measures; semantics; metadata; mapping; OWL ontologies; semantic enrichment; semantic similarity; ontology concepts.

DOI: 10.1504/IJMSO.2013.054179

International Journal of Metadata, Semantics and Ontologies, 2013 Vol.8 No.1, pp.65 - 74

Received: 11 Jul 2012
Accepted: 04 Jan 2013
Published online: 14 Oct 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article