An a Priori Approach for Automatic Integration of Heterogeneous and Autonomous Databases

Data integration is the process that gives users access to multiple data sources through queries against a global schema. The semantic heterogeneity has been identified as the most important and toughest problem when integrating various sources. The mapping between the global schema and local schemas was done manually in the first generation of integrated systems, when ontologies are not used to make explicit data meaning. It is semi automatic when ontologies and ontology mappings are defined at integration level. In this paper, we propose a fully automatic integration approach based on ontologies. It supposes that each data source contains a formal ontology that references a shared ontology. The relationships between each local ontology and the shared ontology are defined at the database design time and also embedded in each source. We assume that a domain ontology exists, but each source may extend it by adding new concepts and properties. Our approach is currently prototyped in various environments: OODB, ORDB, and RDB.