Representing Problem-Solving for Knowledge Refinement

Knowledge refinementtools seek to correct faulty knowledgebased systems (KBSs)by identifying and repairing potentially faulty rules. Thegoal of the KRuSTWorks project is to provide a source of refinement componentsfrom whichspecialised refinementtools tailored to the needs of a range of KBSs are built. A core refinement algorithm reasons about the knowledgethat has been applied, but this approach demands general knowledge structures to represent the reasoning of a particular problemsolving episode. This paper investigates somecomplex formsof rule interaction and defines a knowledgestructure encompassing these. The approach has been applied to KBSs built in four shells andis demonstrated on a small examplethat incorporates someof the complexity foundin real applications.

Authors

• Susan Craw
• Robin Boswell
• The Robert Gordon University

Keywords

No keywords are indexed for this paper.