Abstract
The use of a conceptual model or an ontology over data sources has been shown to be necessary to overcome many important database problems (for a survey see [1]). Since ontologies provide a conceptual view of the application domain, the recent trend to employ such ontologies for navigational (and reasoning) purposes when accessing the data gives additional motivation for the problem of extracting the ontology from database schema [2]. When such an ontology exists, modelling the relation between the data sources and an ontology is a crucial aspect in order to capture the semantics of the data.