This paper presents a specifically database-inspired approach (called DOGMA) for engineering formal ontologies, implemented as shared resources used to express agreed formal semantics for a real world domain. Our method-ology aims to addresses several related issues, such as (a) the scalability of building and sharing ontologies; (b) the maximization of knowledge reusability;(c) the design and engineering process, that also simplifies building and manag-ing ontologies; (d) the coexistence of several rule systems and ontology languages around a same ontology; and (e) the reconcile of the need to represent semantics independently from language with the need to create and use proc-esses entirely rooted and described in (natural) language. We first define formal ontologies in a logic sense, i.e. as "representationless" mathematical objects that form the range of a classical interpretation mapping from a first order language (sometimes called a conceptual schema, and assumed to lexically represent an application), to a set of possible (“plausible”) conceptualizations of the real world domain. We then give a database-inspired "view" on implementations of ontologies seen as resources. Following common model-theoretic database practice we decompose such resources into ontology bases and into of their ex-plicit so-called ontological commitments. Such architecture allows to make the latter (crucial) notion explicit as a separate layer, with concrete and dedicated services, mediating between the ontology base and the application instances that commit to the ontology. We claim it also leads to methodological approaches that naturally extend database modeling theory and practice, and so may in turn lead to scalable solutions for ontology-based systems. We discuss examples of the DOGMA implementation of the ontology base server and commitment server
展开▼
