Survey on two-step fuzzy rule interpolation methods

Abstract

Fuzzy Rule Interpolation (FRI) is a wide applied tool for fuzzy inference when the rule base is sparse, i.e. it does not ensure a full coverage of the input space, and the compositional reasoning methods are not able to produce a proper output in case of each possible observation. This paper surveys a relevant group of FRI techniques called two-step methods that follow the concepts of the generalized methodology of fuzzy rule interpolation (GM) introduced by Baranyi, Kóczy and Gedeon.