Representing Problem-Solving for Knowledge Refinement

Knowledge refinement tools seek to correct faulty knowledge based systems (KBSs) by identifying and repairing potentially faulty rules. The goal of the KRuSTWorks project is to provide a source of refinement components from which specialised refinement tools tailored to the needs of a range of KBSs are built. A core refinement algorithm reasons about the knowledge that has been applied, but this approach demands general knowledge structures to represent the reasoning of a particular problem solving episode. This paper investigates some complex forms of rule interaction and defines a knowledge structure encompassing these. The approach has been applied to KBSs built in four shells and is demonstrated on a small example that incorporates some of the complexity found in real applications.