Abstract
This paper proposes a linguistic modeling method based on a weighted fuzzy rule base and the associated learning algorithm. The fuzzy reference sets and the rule base are simultaneously identified from numeric data in opposition to many other linguistic methods that divide the identification problem into two separate subtasks. No assumption is made on the number of reference sets that may be irregularly distributed according to the training set. Two numeric examples are presented, the first one concerns function approximation and the second one deals with the prediction of Mackey-Glass time series.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have