A hybrid ontology approach for information integration

Information sharing from heterogeneous data sources (different forms and representations of data) in a distributed information system can be challenging because of semantic conflicts and contradictions (different intended meaning of terms in context or application). Various forms of ontology have been used to capture and represent information and semantic meaning. A hybrid approach is presented in this work that combines a global ontology for providing a shared vocabulary for the specification of the semantics of knowledge in a domain, along with local ontologies that describe structures of multiple data sources. A procedure is provided for mapping local ontologies to the global ontology by quantifying relationships between classes and identifying groups of classes with a clustering method. The approach for information integration is demonstrated with an example based on a product obsolescence problem. Ontologies and rules of identifying relationships are represented with OWL (Web Ontology Language) and SWRL (Semantic Web Rule Language) and realized using the ontology editor Protégé. With the application of the hybrid ontology approach, new sources can be easily added with little modification in mappings or in ontologies. As a result, a unified view of data is provided, eliminating semantic conflicts.