Deductive-ER: deductive entity-relationship data model and its data language

Abstract

The entity-relationship (ER) data model is a well known semantic data model in database design. The paper develops a deductive-ER data model, Deductive-ER, which applies the ER approach to the design of deductive databases. Based on such a model, a deductive-ER data language, called Deductive-ER, is constructed, which provides capabilities of definition and manipulation of extensional database components, intensional database components (including virtual entities, virtual relationships, and virtual attributes), hybrid components, generalization hierarchies, and integrity constraints. It supports complex data objects, including tuple-valued, list-valued, text-valued, set-valued, and null-valued attributes. Moreover, it allows recursive definitions. Deductive-ER integrates two different flavoured languages: an ER-based conventional data language and a Horn-clause-based logic data language. It takes advantage of both and provides a simple and powerful interface. The implementation of Deductive-ER is also discussed.