Managing the Mappings between Domain Ontologies and Database Schemas when Formulating Relational Queries

In recent years, the tremendous increase in the use of medical knowledge-discovery and decision-support applications has often required clinical researchers to write complex database queries. The users of these data analysis systems are normally unaware of the semantic relationships between the concepts stored in a database. In order to provide automated query formulation services, some mechanism for generating queries is required. In this regard, as reported in [1], domain ontologies can be used to formulate relational database queries in order to simplify the data access of the underlying data sources. However, the provision of such a query generation facility requires managing complex mappings between domain ontologies and relational data sources. In this regard, this paper discusses our approach to define mappings between domain ontologies and database schemas to support the ontology assisted relational query formulation process. This approach has been applied to the integrated medical database schema of the EU funded Health-e-Child (HeC) project