Case-based reasoning systems solve new problems by retrieving and adapting the solutions to similar previously solved problem cases, and problem solving success depends critically on the quality of these cases. However, it is only relatively recently that researchers have begun to investigate ways of supporting the case-base authoring process in order to improve the quality of the resulting case-bases. In previous work we have argued that visualisation techniques have a valuable role to play in supporting a case-base author by helping the author to understand the structure of the evolving case-base. In this paper we compare and evaluate the 2D and 3D visualisation methods developed as part of the CASCADE case-based reasoning shell. We show that these visualisation methods are capable of producing accurate visual representations of complex case-bases.
展开▼
