Evolutionary Strategies for Generation of Fuzzy Rule Bases: a Local Approach

D. Spiegel; T. Sudkamp

doi:10.1109/SSST.2001.918540

Evolutionary Strategies for Generation of Fuzzy Rule Bases: a Local Approach

D. Spiegel, T. Sudkamp

Computer Science and Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Fuzzy rule bases provide a tool for modeling complex systems and approximating functions. Originally, heuristic analysis by experts was used to produce fuzzy models. Recently, algorithms have been developed to produce models from training data. In this research, two general approaches for evolutionary generation of fuzzy rules are identified and compared: global and local reproduction. Global reproduction, which is the standard approach, considers an entire rule base in performing fitness evaluation and regeneration. The local approach considers a series of independent evolutionary selections and produces a model by combining the localized results. An experimental suite has been developed to compare the effectiveness of the approaches in generating models. The parameters considered include the size office training set and the number of rules.

Original language	English
Title of host publication	Proceedings of the 33rd Southeastern Symposium on System Theory (Cat. No.01EX460)
Publisher	IEEE
Pages	325-329
Number of pages	5
ISBN (Print)	0-7803-6661-1
DOIs	https://doi.org/10.1109/SSST.2001.918540
State	Published - 2001
Event	33rd Southeastern Symposium on System Theory, SSST 2001 - Athens, United States Duration: Mar 18 2001 → Mar 20 2001

Conference

Conference	33rd Southeastern Symposium on System Theory, SSST 2001
Country/Territory	United States
City	Athens
Period	3/18/01 → 3/20/01

ASJC Scopus Subject Areas

Control and Systems Engineering
General Mathematics

Keywords

Algorithm design and analysis
Clustering algorithms
Computer science
Fuzzy sets
Fuzzy systems
Quantization
Takagi-Sugeno-Kang model
Training data

Disciplines

Computer Sciences
Mathematics
Engineering

Access to Document

10.1109/SSST.2001.918540License: Unspecified

Cite this

@inproceedings{c7ff7a663d784d37bc2a36e1693f0e20,

title = "Evolutionary Strategies for Generation of Fuzzy Rule Bases: a Local Approach",

abstract = " Fuzzy rule bases provide a tool for modeling complex systems and approximating functions. Originally, heuristic analysis by experts was used to produce fuzzy models. Recently, algorithms have been developed to produce models from training data. In this research, two general approaches for evolutionary generation of fuzzy rules are identified and compared: global and local reproduction. Global reproduction, which is the standard approach, considers an entire rule base in performing fitness evaluation and regeneration. The local approach considers a series of independent evolutionary selections and produces a model by combining the localized results. An experimental suite has been developed to compare the effectiveness of the approaches in generating models. The parameters considered include the size office training set and the number of rules.",

keywords = "Algorithm design and analysis, Clustering algorithms, Computer science, Fuzzy sets, Fuzzy systems, Quantization, Takagi-Sugeno-Kang model, Training data",

author = "D. Spiegel and T. Sudkamp",

note = "Publisher Copyright: {\textcopyright} 2001 IEEE.; 33rd Southeastern Symposium on System Theory, SSST 2001 ; Conference date: 18-03-2001 Through 20-03-2001",

year = "2001",

doi = "10.1109/SSST.2001.918540",

language = "English",

isbn = "0-7803-6661-1",

pages = "325--329",

booktitle = "Proceedings of the 33rd Southeastern Symposium on System Theory (Cat. No.01EX460)",

publisher = "IEEE",

}

TY - GEN

T1 - Evolutionary Strategies for Generation of Fuzzy Rule Bases

T2 - 33rd Southeastern Symposium on System Theory, SSST 2001

AU - Spiegel, D.

AU - Sudkamp, T.

PY - 2001

Y1 - 2001

N2 - Fuzzy rule bases provide a tool for modeling complex systems and approximating functions. Originally, heuristic analysis by experts was used to produce fuzzy models. Recently, algorithms have been developed to produce models from training data. In this research, two general approaches for evolutionary generation of fuzzy rules are identified and compared: global and local reproduction. Global reproduction, which is the standard approach, considers an entire rule base in performing fitness evaluation and regeneration. The local approach considers a series of independent evolutionary selections and produces a model by combining the localized results. An experimental suite has been developed to compare the effectiveness of the approaches in generating models. The parameters considered include the size office training set and the number of rules.

AB - Fuzzy rule bases provide a tool for modeling complex systems and approximating functions. Originally, heuristic analysis by experts was used to produce fuzzy models. Recently, algorithms have been developed to produce models from training data. In this research, two general approaches for evolutionary generation of fuzzy rules are identified and compared: global and local reproduction. Global reproduction, which is the standard approach, considers an entire rule base in performing fitness evaluation and regeneration. The local approach considers a series of independent evolutionary selections and produces a model by combining the localized results. An experimental suite has been developed to compare the effectiveness of the approaches in generating models. The parameters considered include the size office training set and the number of rules.

KW - Algorithm design and analysis

KW - Clustering algorithms

KW - Computer science

KW - Fuzzy sets

KW - Fuzzy systems

KW - Quantization

KW - Takagi-Sugeno-Kang model

KW - Training data

UR - http://www.scopus.com/inward/record.url?scp=84949967995&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949967995&partnerID=8YFLogxK

UR - https://corescholar.libraries.wright.edu/cse/430

U2 - 10.1109/SSST.2001.918540

DO - 10.1109/SSST.2001.918540

M3 - Conference contribution

SN - 0-7803-6661-1

SP - 325

EP - 329

BT - Proceedings of the 33rd Southeastern Symposium on System Theory (Cat. No.01EX460)

PB - IEEE

Y2 - 18 March 2001 through 20 March 2001

ER -

Evolutionary Strategies for Generation of Fuzzy Rule Bases: a Local Approach

Abstract

Conference

ASJC Scopus Subject Areas

Keywords

Disciplines

Access to Document

Other files and links

Cite this