Choice of implication functions to reduce uncertainty in interval type-2 fuzzy inferences

Abstract

Selection of implication function is a well-known problem in Type-1 fuzzy reasoning. Several comparison of type-1 implications have been reported using set of (nine) standard axioms. This paper attempts to select the most efficient implication function that results in minimum uncertainty in the interval type-2 inference. An analysis confirms that Lukasiewicz-1/Lukasiewicz-2 membership function is most efficient in the present context. © Springer International Publishing Switzerland 2014.