Some methods to model fuzzy systems for inference purposes

https://doi.org/10.1016/S0888-613X(96)00136-3Get rights and content
Under an Elsevier user license
open archive

Abstract

We present different techniques of fuzzy rule generation using the information we can obtain from the fuzzy clustering of a set of data which describe the behavior of a given system. The methods all try to obtain a first model of the consisted system that is good enough to serve as a first approximation for inference purposes. Thus, it is important that the methods should be as simple as possible but with great approximate power.

Keywords

fuzzy clustering
unsupervised learning
fuzzy modeling

Cited by (0)

This work has been partially supported by CICYT project TIC95-1019