Ounissa Larab and Aïcha-Nabila Benharkat
The success of schema integration in multidatabase systems relies heavily on the determination of complete and refined correspondence relationships between them. So, the candidate schemas to be integrated must be rich and precise semantically, i.e., each of their data elements ~ must be sufficiently defined for to be distinguished from others or identified to some of them. Our schema integration method for federated databases is a terminological reasoning-based approach. It deals with the integration of terminologies that translate the export schemas (parts of DB2 schemas which participate to the federation). In spite of it’s semantic power, the terminological formalism can’t express entirely and precisely the real world semantics of schema data elements. To achieve this goal, we think that it is necessary to define refined and complete correspondence relationships between terms. Then, we propose to extend their semantics by a set of additional implicit knowledge which is not represented neither at the description level of terms nor at their extension level. It expresses semantic properties in connection with either local or global context of terminologies that participate to the federation.