Ontology Driven Development of Rule-Based Expert Systems

Abstract

An ontology-driven development process of rule-based expert systems and knowledge bases is described. The process is based on a redefined approach, namely, a model-driven development (MDD). The main feature of the proposed process is the use of ontology as a computation-independent model (CIM). At the same time, ontology is created by analyzing the conceptual models of a subject domain, in particular, UML class diagrams presented in XML-based formats. The Rule Visual Modeling Language (RVML) is used as the extension of a UML for the development of a platform-independent and platform-specific models, and also the C language integrated production system (CLIPS) is used as the target platform. The formalized descriptions of the approach stages and model transformations are considered. The approach proposed allows one: to eliminate programming errors through the automatic code generation; to reduce a time of identification, conceptualization, and formalization stages due to the use of ontologies. The Knowledge Base Development System (KBDS) and the Personal Knowledge Base Designer (PKBD) implement processes and algorithms described and they intend for the rapid development of prototypes of rule-based expert systems and knowledge bases.