Title

Fuzzy Regression by Fuzzy Number Neural Networks

Abstract

In this paper, we describe a method for nonlinear fuzzy regression using neural network models. In earlier work, strong assumptions were made on the form of the fuzzy number parameters: symmetric triangular, asymmetric triangular, quadratic, trapezoidal, and so on. Our goal here is to substantially generalize both linear and nonlinear fuzzy regression using models with general fuzzy number inputs, weights, biases, and outputs. This is accomplished through a special training technique for fuzzy number neural networks. The technique is demonstrated with data from an industrial quality control problem.

Department(s)

Electrical and Computer Engineering

Keywords and Phrases

Back Propagation; Fuzzy Regression; Neural Networks

International Standard Serial Number (ISSN)

0165-0114

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2000 Elsevier, All rights reserved.


Share

 
COinS