Usually, fuzzy rules contain in the antecedent propositions that restrict a variable to a fuzzy value by means of an equal-to predicate. We propose to improve the interpretability of fuzzy models by extending the syntax of their rules. With this aim, on one hand, new predicates are considered in the rule antecedents and, on the other hand, rules can be associated with exceptions that modify the output of those rules in a region of their covered input space. The method stems from an initial fuzzy model described with the usual fuzzy rules and uses an ACO algorithm to search the optimal set of extended rules that describes this model. © 2007 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Carmona, P., & Castro, J. L. (2007). An ant colony optimization plug-in to enhance the interpretability of fuzzy rule bases with exceptions. Advances in Soft Computing, 41, 436–444. https://doi.org/10.1007/978-3-540-72432-2_44
Mendeley helps you to discover research relevant for your work.