Towards Automated Reasoning on ORM Schemes Mapping ORM into the DLRidf Description Logic

The goal of this article is to formalize Object Role Modeling (ORM) using the DLR description logic. This would enable automated reasoning on the formal properties of ORM diagrams, such as detecting constraint contradictions and implications. In addition, the expressive, methodological, and graphical capabilities of ORM make it a good candidate for use as a graphical notation for most description logic languages. In this way, industrial experts who are not IT savvy will still be able to build and view axiomatized theories (such as ontologies, business rules, etc.) without needing to know the logic or reasoning foundations underpinning them. Our formalization in this paper is structured as 29 formalization rules, that map all ORM primitives and constraints into DLR, and 2 exceptions of complex cases. To this end, we illustrate the implementation of our formalization as an extension to DogmaModeler, which automatically maps ORM into DIG and uses Racer as a background reasoning engine to reason about ORM diagrams. 1 Motivation and Background This article proposes to formalize ORM (Object Role Modeling [8]) using the DLR description logic. This would enable automated reasoning to be carried out on the formal properties of ORM diagrams, such as detecting constraint contradictions and implications. In addition, the expressive, methodological, and graphical power of ORM make it a good candidate for use as a graphical notation for most description logic languages. With this, non-IT trained industrial experts will be able to build axiomatized theories (such as ontologies, business rules, etc.) in a graphical manner, without having to know the underpinning logic or foundations. ORM is a conceptual modeling method that allows the semantics of a universe of discourse to be modeled at a highly conceptual level and in a graphical manner. ORM has been used commercially for more than 30 years as a database modeling methodology, and has recently becoming popular not only for ontology engineering but also as a graphical notation in other areas such as the modeling of business rules, XML-Schemes, data warehouses, requirements engineering,web forms, etc1. The author is currently moving from Brussels to Nicosia and soon will be affiliated only with the university of Cyprus. 1 Many commercial and academic tools that support ORM solutions are available, including the ORM solution within Microsoft’s Visio for Enterprise Architects, VisioModeler, NORMA, CaseTalk, Infagon, and DogmaModeler. DogmaModeler and its support for ontology engineering will be presented later in this paper. C. Parent et al. (Eds.): ER 2007, LNCS 4801, pp. 181–197, 2007. c © Springer-Verlag Berlin Heidelberg 2007