Facilitating connectivity in composite information systems

Timely access to multiple disparate databases which were independently developed and administered to produce composite information has become increasingly critical for organizations to gain competitive advantage. However, many inter-database problems such as inconsistency, ambiguity, and contradiction remain unresolved. This paper presents an approach for resolving these problems. The techniques employed in this approach include schema integration, inter-database tables, attribute subsetting, object hierarchies, and heuristic rules. Schema integration techniques resolve the incompatibilities among the databases at the schema level. Inter-database tables resolve the semantic inconsistency and concept granularity at the instance value level. The inter-database instance identification table identifies an instance across databases. Object hierarchies represent schemata as well as instances. Finally, heuristic rules are used to facilitate the construction of the inter-database instance identification table and the production of composite information.