Ontology-based Mediation with Quality Criteria

Abstract

This paper presents a semantic system named OntMed for an ontology-based data integration of heterogeneous data sources to achieve interoperability between heterogeneous data sources. Our system is based on the quality criteria (consistency, completeness and conciseness) for building the reliable analysis contexts to provide an accurate unified view of data to the end user. The generation of an error-free global analysis context with the semantic validation of initial mappings generates accuracy, and provides the means to access and exchange information in semantically sound manner. In addition, data integration in this way becomes more practical for dynamic situations and helps decision maker to work within more consistent and reliable virtual data warehouse. We also discuss our successful participation in the Ontology Alignment for Query Answering (OA4QA) track at OAEI 2015 campaign, where our system (DKP-AOM) has performed fair enough and became one of only matchers whose alignments allowed answering all the queries of the evaluation.