Tuning fuzzy aggregation operators using fuzzy meta-rules

Tuning or choosing a fuzzy aggregation operator to achieve the best performance for a fuzzy system can be difficult (e.g. in some systems the best performing operator may be the min operator whereas in others it may be the algebraic product). This paper discusses how the prognostic error measure can be used to compare the differences in behavior of various operators. It is then shown how one operator can use fuzzy meta-rules to approximate the behavior of another operator. Thus one can tune an operator to behave similarly to another operator