Design and tuning of fuzzy if-then rules for automatic classification
Alexander P. Rotshtein, D. I. Katel’nikov · 2002
We propose an approach to the design and tuning of fuzzy rules for automatic classification decision making. This approach is based upon the finding of the weights of fuzzy if-then rules and the shapes of membership functions that minimize the difference between real (desired) and inferred (theoretical) classes of decisions. The problem of fuzzy model tuning is stated as a classical mathematical optimization problem.