Recent work in Artificial Intelligence is exploring the use of formal ontologies as a way of specifying content-specific agreements for the sharing and reuse of knowledge among software entities. We take an engineering perspective on the development of such ontologies. Formal ontologies are viewed as designed artifacts, formulated for specific purposes and evaluated against objective design criteria. We describe the role of ontologies in supporting knowledge sharing activities, and then present a set of criteria to guide the development of ontologies for these purposes. We show how these criteria are applied in case studies from the design of ontologies for engineering mathematics and bibliographic data. Selected design decisions are discussed, and alternative representation choices and evaluated against the design criteria.

The paper discusses an approach which allows the specification of a knowledge-based system (kbs) at several levels. The Knowledge Acquisition and Representation Language KARL combines a description of a kbs at the conceptual level supported by graphical modelling primitives with a description at a formal and executable level. Therefore, a KARLspecification can be used as a means for communication between expert and knowledge engineer as well as an intermediate representation, closing the conceptual gap between an informal specification and an implementation of a kbs. In the paper, KARL is mainly discussed as a graphical modelling language. Introduction There is a clear need for combining formal and executable specification of a kbs with descriptions at a high conceptual level supported by graphical representation primitives. In the paper, we discuss the Knowledge Acquisition and Representation Language KARL, which integrates formal specifications with a graphical modelling language...