An Ontology-Based Data Integration System with Dynamic Concept Mapping and Plug-In Management

The existing data integration systems which are not suitable for the changeable needs of enterprises become more and more complex and unfeasible with the increasing heterogeneous data sources. To solve the problem, this paper proposes a non-instance ontology-based approach of concepts mapping and an approach of plug-in management. Ontology mapping is used to provide basic data to integrate the semantic data sources into the system. The plug-in management makes the bundle of distributed data source service handle its life-cycle and enables the data queries bind the relative services dynamically. Empirical results show that the method of the concepts mapping can maintain the accuracy of positioning the concepts during the process of the integration, and the plug-in service management can handle the queries dynamically. An ontology-based data integration system with dynamic concept mapping and plug-in management shows that the novel idea is feasible and dynamical for the changeable needs of the system.