Article Preview
Top2. Background
Conceptual modeling is the activity of formally describing some aspects of the physical and social world around us for purposes of understanding and communication (Mylopoulos, 1992). The more common uses of conceptual models in the IS field are to: (1) facilitate communications between users and analysts, (2) support the analysts’ understanding of the domain, (3) serve as the basis for design and implementation of IS, and (4) record design rationales (Kung & Solvberg, 1986). While conceptual models provide input for design, they do not represent the IS artefact. In particular, conceptual models are different than semantic data models. In particular, conceptual models are created for studying a business, while semantic data models are created for designing a database.
While an IS ontology defines a set of concepts, a conceptual model uses concepts to represent a specific domain. Conceptual models are created using modeling grammars comprising constructs for representing domain phenomena, and rules for combining these constructs (Shanks et al., 2003). There are at least two reasons why it might be advantageous to use an ontology language as a conceptual modeling grammar. First, using a formalized ontology language can provide for including the semantics of domain concepts as part of the conceptual model. Second, ontology language statements are intended to be processed by software applications and can be subject to automated reasoning. Hence, conceptual models represented in ontology languages can be subject to automated processing, in particular to verification beyond what graphical representation affords.
However, ontology language constructs do not have the domain semantics required from conceptual models. We propose that since philosophical theories of ontology can represent domain phenomena (Shanks et al., 2003; Wand & Weber, 2002), such theories can guide the use of ontology languages for conceptual modeling.