A New Approach to Rule Learning Based on Fusion of Fuzzy Logic and Neural Networks

Rui-Ping LI; Masao MUKAIDONO

A New Approach to Rule Learning Based on Fusion of Fuzzy Logic and Neural Networks

Rui-Ping LI, Masao MUKAIDONO

Full Text Views

0

Cite this

Summary :

A new method is developed to generate fuzzy rules from numerical data. This method consists of two algorithms: Algorithm 1 is used to identify structures of the given data set, that is, the optimal number of rules of system; Algorithm 2 is used to identify parameter of the used model. The former is belonged to unsupervised learning, and the latter is belonged to supervised learning. To identify parameters of fuzzy model, we developed a neural network which is referred to as Unsymmetrical Gaussian Function Network (UGFN). Unlike traditional fuzzy modelling methods, in the present method, a) the optimal number of rules (clusters) is determinde by input-output data pairs rather than by only output data as in sugeno's method, b) parameter identification of ghe present model is based on a like-RBF network rather than backpropagation algorithm. Our method is simple and effective because it integrates fuzzy logic with neural networks from basic network principles to neural architecture, thereby establishing an unifying framework for different fuzzy modelling methods such as one with cluster analysis or neural networks and so on.

Publication: IEICE TRANSACTIONS on Information Vol.E78-D No.11 pp.1509-1514

Publication Date: 1995/11/25

Publicized

Online ISSN

DOI

Type of Manuscript: PAPER

Category: Bio-Cybernetics and Neurocomputing

Cite this

Copy

Rui-Ping LI, Masao MUKAIDONO, "A New Approach to Rule Learning Based on Fusion of Fuzzy Logic and Neural Networks" in IEICE TRANSACTIONS on Information, vol. E78-D, no. 11, pp. 1509-1514, November 1995, doi: .
Abstract: A new method is developed to generate fuzzy rules from numerical data. This method consists of two algorithms: Algorithm 1 is used to identify structures of the given data set, that is, the optimal number of rules of system; Algorithm 2 is used to identify parameter of the used model. The former is belonged to unsupervised learning, and the latter is belonged to supervised learning. To identify parameters of fuzzy model, we developed a neural network which is referred to as Unsymmetrical Gaussian Function Network (UGFN). Unlike traditional fuzzy modelling methods, in the present method, a) the optimal number of rules (clusters) is determinde by input-output data pairs rather than by only output data as in sugeno's method, b) parameter identification of ghe present model is based on a like-RBF network rather than backpropagation algorithm. Our method is simple and effective because it integrates fuzzy logic with neural networks from basic network principles to neural architecture, thereby establishing an unifying framework for different fuzzy modelling methods such as one with cluster analysis or neural networks and so on.
URL: https://global.ieice.org/en_transactions/information/10.1587/e78-d_11_1509/_p

Copy

@ARTICLE{e78-d_11_1509,
author={Rui-Ping LI, Masao MUKAIDONO, },
journal={IEICE TRANSACTIONS on Information},
title={A New Approach to Rule Learning Based on Fusion of Fuzzy Logic and Neural Networks},
year={1995},
volume={E78-D},
number={11},
pages={1509-1514},
abstract={A new method is developed to generate fuzzy rules from numerical data. This method consists of two algorithms: Algorithm 1 is used to identify structures of the given data set, that is, the optimal number of rules of system; Algorithm 2 is used to identify parameter of the used model. The former is belonged to unsupervised learning, and the latter is belonged to supervised learning. To identify parameters of fuzzy model, we developed a neural network which is referred to as Unsymmetrical Gaussian Function Network (UGFN). Unlike traditional fuzzy modelling methods, in the present method, a) the optimal number of rules (clusters) is determinde by input-output data pairs rather than by only output data as in sugeno's method, b) parameter identification of ghe present model is based on a like-RBF network rather than backpropagation algorithm. Our method is simple and effective because it integrates fuzzy logic with neural networks from basic network principles to neural architecture, thereby establishing an unifying framework for different fuzzy modelling methods such as one with cluster analysis or neural networks and so on.},
keywords={},
doi={},
ISSN={},
month={November},}

Copy

TY - JOUR
TI - A New Approach to Rule Learning Based on Fusion of Fuzzy Logic and Neural Networks
T2 - IEICE TRANSACTIONS on Information
SP - 1509
EP - 1514
AU - Rui-Ping LI
AU - Masao MUKAIDONO
PY - 1995
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E78-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 1995
AB - A new method is developed to generate fuzzy rules from numerical data. This method consists of two algorithms: Algorithm 1 is used to identify structures of the given data set, that is, the optimal number of rules of system; Algorithm 2 is used to identify parameter of the used model. The former is belonged to unsupervised learning, and the latter is belonged to supervised learning. To identify parameters of fuzzy model, we developed a neural network which is referred to as Unsymmetrical Gaussian Function Network (UGFN). Unlike traditional fuzzy modelling methods, in the present method, a) the optimal number of rules (clusters) is determinde by input-output data pairs rather than by only output data as in sugeno's method, b) parameter identification of ghe present model is based on a like-RBF network rather than backpropagation algorithm. Our method is simple and effective because it integrates fuzzy logic with neural networks from basic network principles to neural architecture, thereby establishing an unifying framework for different fuzzy modelling methods such as one with cluster analysis or neural networks and so on.
ER -