Author(s)

K. Takezawa

Abstract

A new method of acquiring fuzzy rules using observational data is proposed. The Turbo Algorithm, which has attracted much attention as an efficient procedure for deriving the nonparametric functions from data, plays a major role in this algorithm. Since the algorithm selects the best combination of rules from various candidates automatically, reliable estimation can be achieved through only a small number of rules. To confirm its validity, this method is applied to the analysis of environmental effects on rice production. 1. Introduction Identification of the relationship between input and output using observed data is a useful technique for estimation, prediction and control. To match the recent progress in computer technology, flexible algorithms for this

Keywords