Global Schema Generation Using Formal Ontologies

This paper deals with the problem of handling semantic heterogeneity during schema integration. Semantics refer to the meaning of data in contrast to syntax, which solely defines the structure of schema elements. We focus on the part of semantics related to the meanings of terms used to name schema elements. Our approach does not rely on the names of the schema elements or the structure of the schema. Instead, we present an approach based on formal ontologies presented in a logical language for integrating schemas to produce global schemas. Semantic similarity relations between definitions in formal ontologies are defined and used for merging formal ontologies. We show how similarity relations are discovered by a reasoning system using a higher level ontology. The result of the merging process is used for schema integration. The result of schema integration can be used as a global schema in a tightly coupled federated database. Afterwards, we illustrate how the produced global schema can help for the mapping of data elements.