Skip to Main Content
Ontologies are becoming efficient models for information representation and storage. They facilitate treatment and knowledge management through AI techniques by offering the potential of integrating a large quantity of information via what we call “Ontology Merging”. Beforehand, each data source or DB may be the object of an ontology construction. Our contribution is to design a syntactico-semantic algorithm for automatic ontology merging. It combines syntactic and semantic measures for identifying similar concepts that will be merged in a signal one in the resulting merged ontology. Hence, synonym and homonym problems of the syntactic measures can be solved. The syntactic part of the algorithm is based on calculating the distance between the two compared concepts while the semantic one is based on the extensional models of the source ontologies from WordNet which constitute the basis of semantic similarity measure between the synsets of the two concepts in question. After combining the two results, similar concepts are merged in a single one.