Fuzzy temporal sequence processing by recurrent neural fuzzy network

Author

Juang, Chia-Feng ; Ku, Shiuan-Jiun ; Huang, Hao-Jung

Author_Institution

Dept. of Electr. Eng., Nat. Chung-Hsing Univ., Taichung, Taiwan

Volume

6

fYear

2004

fDate

10-13 Oct. 2004

Firstpage

5847

Abstract

A fuzzified TSK-type recurrent neural fuzzy network (FTRNFN) for handling fuzzy temporal information is proposed in this paper. The inputs and outputs of FTRNFN are fuzzy patterns represented by Gaussian or isosceles triangular membership functions. In structure, FTRNFN is a recurrent fuzzy network constructed from a series of recurrent fuzzy if-then rules with TSK-type consequent parts. The recurrent property of FTRNFN enables it to deal with fuzzy patterns with temporal context. There are no rules in FTRNFN initially; they are constructed on-line by concurrent structure and parameter learning. The ability of TRFNFN is verified from a two-dimensional fuzzy temporal sequence prediction problem.

Keywords

Gaussian processes; fuzzy neural nets; fuzzy set theory; recurrent neural nets; Gaussian representation; fuzzy patterns; fuzzy temporal information handling; isosceles triangular membership functions; parameter learning; recurrent neural fuzzy network; two-dimensional fuzzy temporal sequence prediction problem; Fuzzy neural networks; Fuzzy sets; Fuzzy systems; Humans; Multi-layer neural network; Neural networks; Process design; Recurrent neural networks;

fLanguage

English

Publisher

ieee

Conference_Titel

Systems, Man and Cybernetics, 2004 IEEE International Conference on

ISSN

1062-922X

Print_ISBN

0-7803-8566-7

Type

conf

DOI

10.1109/ICSMC.2004.1401128

Filename

1401128