Integration of relational databases into OWL knowledge bases:: demonstration of the DBOM system

In this paper, we present a system which enables the integration of data stored in relational databases into a Semantic Web compliant knowledge base. In this context, our system, named DBOM, proposes a solution to the impedance mismatch problem by proposing a mapping language that allows to specify how to transform data retrieved from tuples of the databases into objects of the target knowledge base. DBOM also handles inconsistencies that may be caused by the execution of uncertain mappings. The proposed solution is based on setting preferences over the views and attributes of the views of the mappings. We demonstrate DBOM in the context of a medical informatics application that requires the integration of drug related information into an OWL knowledge base.