Fuzzy Regression by Fuzzy Number Neural Networks
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.
J. Dunyak and D. C. Wunsch, "Fuzzy Regression by Fuzzy Number Neural Networks," Fuzzy Sets and Systems, vol. 112, no. 3, pp. 371 - 380, Elsevier, Jun 2000.
The definitive version is available at https://doi.org/10.1016/S0165-0114(97)00393-X
Electrical and Computer Engineering
Keywords and Phrases
Back Propagation; Fuzzy Regression; Neural Networks
International Standard Serial Number (ISSN)
Article - Journal
© 2000 Elsevier, All rights reserved.
01 Jun 2000