Fuzzy models, identification and applications

L. Kóczy, János Botzheim

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

This paper gives a brief overview of fuzzy model identification techniques. The paper discusses how the membership functions of a fuzzy system can be extracted from an input/output data (pattern) set without human interference. There are several methods used for rule extraction known from the literature. The bacterial algorithm is an evolutionary technique that was inspired by the microbial evolution phenomenon. The Levenberg-Marquardt algorithm is an advanced gradient type optimization method that has been developed initially for neural networks and is introduced here for the optimization of the fuzzy rule base. Fuzzy clustering is presented also as another alternative way for the rule extraction. In the part describing the model the fuzzy rule interpolation method and the approach of hierarchical rule bases are introduced. Combining fuzzy rule interpolation with the use of hierarchically structured fuzzy rule bases leads to the reduction of the fuzzy algorithms' complexity. Hierarchical fuzzy modeling by clustering techniques is also introduced in the paper.

Original languageEnglish
Title of host publicationICCC 2005 - IEEE 3rd International Conference on Computational Cybernetics - Proceedings
Pages13-19
Number of pages7
Volume2005
DOIs
Publication statusPublished - 2005
EventICCC 2005 - IEEE 3rd International Conference on Computational Cybernetics - Mauritius, Mauritius
Duration: Apr 13 2005Apr 16 2005

Other

OtherICCC 2005 - IEEE 3rd International Conference on Computational Cybernetics
CountryMauritius
CityMauritius
Period4/13/054/16/05

Fingerprint

Fuzzy rules
Identification (control systems)
Interpolation
Fuzzy clustering
Fuzzy systems
Membership functions
Neural networks

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Kóczy, L., & Botzheim, J. (2005). Fuzzy models, identification and applications. In ICCC 2005 - IEEE 3rd International Conference on Computational Cybernetics - Proceedings (Vol. 2005, pp. 13-19). [1511538] https://doi.org/10.1109/ICCCYB.2005.1511538

Fuzzy models, identification and applications. / Kóczy, L.; Botzheim, János.

ICCC 2005 - IEEE 3rd International Conference on Computational Cybernetics - Proceedings. Vol. 2005 2005. p. 13-19 1511538.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kóczy, L & Botzheim, J 2005, Fuzzy models, identification and applications. in ICCC 2005 - IEEE 3rd International Conference on Computational Cybernetics - Proceedings. vol. 2005, 1511538, pp. 13-19, ICCC 2005 - IEEE 3rd International Conference on Computational Cybernetics, Mauritius, Mauritius, 4/13/05. https://doi.org/10.1109/ICCCYB.2005.1511538
Kóczy L, Botzheim J. Fuzzy models, identification and applications. In ICCC 2005 - IEEE 3rd International Conference on Computational Cybernetics - Proceedings. Vol. 2005. 2005. p. 13-19. 1511538 https://doi.org/10.1109/ICCCYB.2005.1511538
Kóczy, L. ; Botzheim, János. / Fuzzy models, identification and applications. ICCC 2005 - IEEE 3rd International Conference on Computational Cybernetics - Proceedings. Vol. 2005 2005. pp. 13-19
@inproceedings{54c41a36ad054bd088e88bee309a16e4,
title = "Fuzzy models, identification and applications",
abstract = "This paper gives a brief overview of fuzzy model identification techniques. The paper discusses how the membership functions of a fuzzy system can be extracted from an input/output data (pattern) set without human interference. There are several methods used for rule extraction known from the literature. The bacterial algorithm is an evolutionary technique that was inspired by the microbial evolution phenomenon. The Levenberg-Marquardt algorithm is an advanced gradient type optimization method that has been developed initially for neural networks and is introduced here for the optimization of the fuzzy rule base. Fuzzy clustering is presented also as another alternative way for the rule extraction. In the part describing the model the fuzzy rule interpolation method and the approach of hierarchical rule bases are introduced. Combining fuzzy rule interpolation with the use of hierarchically structured fuzzy rule bases leads to the reduction of the fuzzy algorithms' complexity. Hierarchical fuzzy modeling by clustering techniques is also introduced in the paper.",
author = "L. K{\'o}czy and J{\'a}nos Botzheim",
year = "2005",
doi = "10.1109/ICCCYB.2005.1511538",
language = "English",
isbn = "0780391225",
volume = "2005",
pages = "13--19",
booktitle = "ICCC 2005 - IEEE 3rd International Conference on Computational Cybernetics - Proceedings",

}

TY - GEN

T1 - Fuzzy models, identification and applications

AU - Kóczy, L.

AU - Botzheim, János

PY - 2005

Y1 - 2005

N2 - This paper gives a brief overview of fuzzy model identification techniques. The paper discusses how the membership functions of a fuzzy system can be extracted from an input/output data (pattern) set without human interference. There are several methods used for rule extraction known from the literature. The bacterial algorithm is an evolutionary technique that was inspired by the microbial evolution phenomenon. The Levenberg-Marquardt algorithm is an advanced gradient type optimization method that has been developed initially for neural networks and is introduced here for the optimization of the fuzzy rule base. Fuzzy clustering is presented also as another alternative way for the rule extraction. In the part describing the model the fuzzy rule interpolation method and the approach of hierarchical rule bases are introduced. Combining fuzzy rule interpolation with the use of hierarchically structured fuzzy rule bases leads to the reduction of the fuzzy algorithms' complexity. Hierarchical fuzzy modeling by clustering techniques is also introduced in the paper.

AB - This paper gives a brief overview of fuzzy model identification techniques. The paper discusses how the membership functions of a fuzzy system can be extracted from an input/output data (pattern) set without human interference. There are several methods used for rule extraction known from the literature. The bacterial algorithm is an evolutionary technique that was inspired by the microbial evolution phenomenon. The Levenberg-Marquardt algorithm is an advanced gradient type optimization method that has been developed initially for neural networks and is introduced here for the optimization of the fuzzy rule base. Fuzzy clustering is presented also as another alternative way for the rule extraction. In the part describing the model the fuzzy rule interpolation method and the approach of hierarchical rule bases are introduced. Combining fuzzy rule interpolation with the use of hierarchically structured fuzzy rule bases leads to the reduction of the fuzzy algorithms' complexity. Hierarchical fuzzy modeling by clustering techniques is also introduced in the paper.

UR - http://www.scopus.com/inward/record.url?scp=33749051166&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33749051166&partnerID=8YFLogxK

U2 - 10.1109/ICCCYB.2005.1511538

DO - 10.1109/ICCCYB.2005.1511538

M3 - Conference contribution

AN - SCOPUS:33749051166

SN - 0780391225

SN - 9780780391222

VL - 2005

SP - 13

EP - 19

BT - ICCC 2005 - IEEE 3rd International Conference on Computational Cybernetics - Proceedings

ER -