DocumentCode :
3086253
Title :
A new approach to generate a self-organizing fuzzy neural network model
Author :
Leng, G. ; Prasad, G. ; McGinnity, T.M.
Author_Institution :
Intelligent Syst. Eng. Lab., Univ. of Ulster, Londonderry, UK
Volume :
4
fYear :
2002
fDate :
6-9 Oct. 2002
Abstract :
This paper presents a new approach for creating a self-organizing fuzzy neural network (SOFNN) from training data, to implement the Takagi-Sugeno-Kang (TSK) model. The center vector and the width vector have been introduced in the RBF neurons in the SOFNN. Novel methods of structure learning and parameter learning, based on new adding and pruning techniques and a recursive on-line learning algorithm, are proposed and developed. The proposed methods are very simple and effective and generate a fuzzy neural model with a high accuracy and a very compact structure. Simulation studies based on a pH neutralization process, confirm that the SOFNN has the capability of self-organization, and can determine the structure and parameters of the network automatically without non-linear optimization.
Keywords :
fuzzy neural nets; learning (artificial intelligence); least squares approximations; neural net architecture; self-organising feature maps; SOFNN; Takagi-Sugeno-Kang model; center vector; pH neutralization process; parameter learning; pruning; recursive least squares algorithm; recursive online learning algorithm; self-organizing fuzzy neural network model; simulation; structure learning; training data; width vector; Biological neural networks; Computer networks; Data engineering; Fuzzy neural networks; Input variables; Intelligent networks; Intelligent systems; Neurons; Organizing; Systems engineering and theory;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Systems, Man and Cybernetics, 2002 IEEE International Conference on
ISSN :
1062-922X
Print_ISBN :
0-7803-7437-1
Type :
conf
DOI :
10.1109/ICSMC.2002.1173311
Filename :
1173311
Link To Document :
بازگشت