Developing Knowledge-Based Systems with MIKE

  • Authors:

  • J. Angele;D. Fensel;D. Landes;R. Studer

  • Affiliations:

  • Institute AIFB, University of Karlsruhe, 76128 Karlsruhe, Germany. E-mail: Email: angele@aifb.uni-karlsruhe.de;Institute AIFB, University of Karlsruhe, 76128 Karlsruhe, Germany. E-mail: Email: fensel@aifb.uni-karlsruhe.de;Institute AIFB, University of Karlsruhe, 76128 Karlsruhe, Germany. E-mail: Email: landes@dbag.ulm.DaimlerBenz.COM;Institute AIFB, University of Karlsruhe, 76128 Karlsruhe, Germany. E-mail: Email: studer@aifb.uni-karlsruhe.de

  • Venue:
  • Automated Software Engineering
  • Year:
  • 1998

Quantified Score

Hi-index 0.00

Visualization

Abstract

The paper describes the MIKE (Model-based andIncremental Knowledge Engineering) approach for developingknowledge-based systems. MIKE integrates semiformal and formalspecification techniques together with prototyping into a coherentframework. All activities in the building process of aknowledge-based system are embedded in a cyclic process model. Forthe semiformal representation we use a hypermedia-based formalismwhich serves as a communication basis between expert and knowledgeengineer during knowledge acquisition. The semiformal knowledgerepresentation is also the basis for formalization, resulting in aformal and executable model specified in the Knowledge Acquisitionand Representation Language (KARL). Since KARL is executable, themodel of expertise can be developed and validated by prototyping. Asmooth transition from a semiformal to a formal specification andfurther on to design is achieved because all the descriptiontechniques rely on the same conceptual model to describe thefunctional and nonfunctional aspects of the system. Thus, the systemis thoroughly documented at different description levels, each ofwhich focuses on a distinct aspect of the entire development effort.Traceability of requirements is supported by linking the differentmodels to each other.