Report on the EDBT'02 panel on scientific data integration

Various issues related to scientific data integration are discussed. The goal of data integration is to construct a global description, called global schema, of the data coming from a multitude of heterogeneous sources. Data integration systems generally follow a semantic approach to integration based on the conceptual schemas or metadata of the sources to be integrated and on a middleware data model for a uniform and semantically rich representation of heterogeneous sources. A key issue of the capability to bring data semantics aspects in automated schema matching techniques for semantic data integration.