Simplification of fuzzy rule based systems using orthogonal transformation

Author

Yen, John ; Wang, Liang

Author_Institution

Dept. of Comput. Sci., Texas A&M Univ., College Station, TX, USA

Volume

1

fYear

1997

fDate

1-5 Jul 1997

Firstpage

253

Abstract

It is known that removal of those redundant fuzzy rules from a rule base can result in a more compact fuzzy model with better generalizing ability. In this paper we propose a number of orthogonal transformation based methods which provide new or alternative tools for rule extraction. A common attribute of these methods is that they all work on a truth value matrix and employ some measure index to detect the rules that should be retained and eliminated. The performance of these methods is illustrated using a nonlinear plant modeling example

Keywords

eigenvalues and eigenfunctions; fuzzy systems; generalisation (artificial intelligence); knowledge acquisition; knowledge based systems; least squares approximations; nonlinear systems; singular value decomposition; truth maintenance; eigenvalue decomposition; fuzzy model; fuzzy rule based systems; least squares methods; measure index; nonlinear plant modeling; orthogonal transformation; rule extraction; singular value decomposition; truth value matrix; Computer science; Fuzzy logic; Fuzzy systems; Input variables; Intelligent robots; Intelligent systems; Knowledge based systems; Least squares methods; Singular value decomposition; State-space methods;

fLanguage

English

Publisher

ieee

Conference_Titel

Fuzzy Systems, 1997., Proceedings of the Sixth IEEE International Conference on

Conference_Location

Barcelona

Print_ISBN

0-7803-3796-4

Type

conf

DOI

10.1109/FUZZY.1997.616377

Filename

616377