As an extension of simplified distance-type reasoning method, multi-output distance-type TS (Takagi-Sugeno) reasoning method is proposed in which the consequent part of the fuzzy rule is functions and 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 medical data of liver ailment diagnosis data.
