Generating Fuzzy Membership Functions through a Meta-Function: An Experience Mining Approach

One of the main issues concerning the design of a fuzzy logic system is generation of the membership functions for the linguistic variable values. Regarding this point, in this paper, we present a new approach for generation of FMF for a new user emphasizing the important role of past experiences, which have been generated in past experiences, and the genealogical factors with respect to the given linguistic variable value. In order to generate a new FMF, we have made use of case-based reasoning and in the meantime have introduced the notion of meta-function as a flexible framework for this purpose.