Design of fuzzy inference systems using regression trees
Sook Lim, Sung Chun Kim · 2002
In this paper, we describe a design methodology for fuzzy inference systems based on regression trees. Fuzzy associative memory banks with symmetric triangular fuzzy membership functions are constructed by using sample data statistics of regression tree partitions. We propose a new fuzzy inference method that takes the membership of the average of normalized input features in the antecedent of a production rule.