9P-D-10 Multi-output Simplified Distance-type Reasoning Method and Its Application to Fuzzy Discriminant Analysis for Medical Diagnosis Data

Abstract

As an extension of simplified distance-type reasoning method, multi-output simplified distance-type reasoning method is proposed in which the consequent part of the fuzzy rule is of multi-output type. This paper shows learning algorithms for the fuzzy reasoning method by using the steepest decent method, and compares learning results with the case of ordinary distance-type reasoning method by applying to two kinds of medical data of diabetes and liver ailment diagnosis data.