AKT-R4: Supporting the Exploration of a Knowledge Web

This paper outlines a proposal for an expert system, AKT-R4, which will be the fourth iteration of the Refiner series. Unlike previous iterations, AKTR4 will include a non-deterministic feature which will enable it to better handle noisy datasets by generating category descriptions with ‘fuzzy’ boundaries. The AKT-R4 system will be based on a case-based algorithm focussed on performing a classification task, in particular medical (differential) diagnosis; AKT-R4 will be aimed primarily at medical students and junior doctors (i.e., novice diagnosticians). The AKT-R4 algorithm will incorporate concepts from casebased reasoning, hypothetico-deductive reasoning (HDR) and illness scripts, and will perform diagnosis by means of a diagnostic web, which is an expansion of the illness script concept. This system supports the novice user as they develop and navigate through a medical knowledge web.