Statistical confidence measure of fuzzy logic system performance
J.X. Lee, George Vukovich · 1999
Statistical confidence measures of performance of a fuzzy logic system in identification problems are proposed. Specifically, by using interval estimation techniques of statistics, we present the statistical estimates of confidence limits on the mean of fuzzy logic system output error as well as on the means of prediction errors of individual IF-THEN rules. With these quantitative measures incorporated, the performances of the fuzzy logic system can be more accurately described and analyzed, and the overall system reliability is expected to be enhanced.