Modified Gath-Geva fuzzy clustering for identification of Takagi-Sugeno fuzzy models

J. Abonyi, Robert Babuška, F. Szeifert

Research output: Contribution to journalArticle

236 Citations (Scopus)

Abstract

The construction of interpretable Takagi-Sugeno (TS) fuzzy models by means of clustering is addressed. First, it is shown how the antecedent fuzzy sets and the corresponding consequent parameters of the TS model can be derived from clusters obtained by the Gath-Geva (GG) algorithm. To preserve the partitioning of the antecedent space, linearly transformed input variables can be used in the model. This may, however, complicate the interpretation of the rules. To form an easily interpretable model that does not use the transformed input variables, a new clustering algorithm is proposed, based on the expectation-maximization (EM) identification of Gaussian mixture models. This new technique is applied to two well-known benchmark problems: the MPG (miles per gallon) prediction and a simulated second-order nonlinear process. The obtained results are compared with results from the literature.

Original languageEnglish
Pages (from-to)612-621
Number of pages10
JournalIEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Volume32
Issue number5
DOIs
Publication statusPublished - Oct 2002

Fingerprint

Fuzzy clustering
Identification (control systems)
Fuzzy sets
Clustering algorithms

Keywords

  • Clustering methods
  • Expectation maximization
  • Fuzzy systems
  • Modeling
  • Optimization
  • Regression

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Artificial Intelligence
  • Human-Computer Interaction

Cite this

@article{b8c35fa4ebed4831bdd65a4d9d3efe8b,
title = "Modified Gath-Geva fuzzy clustering for identification of Takagi-Sugeno fuzzy models",
abstract = "The construction of interpretable Takagi-Sugeno (TS) fuzzy models by means of clustering is addressed. First, it is shown how the antecedent fuzzy sets and the corresponding consequent parameters of the TS model can be derived from clusters obtained by the Gath-Geva (GG) algorithm. To preserve the partitioning of the antecedent space, linearly transformed input variables can be used in the model. This may, however, complicate the interpretation of the rules. To form an easily interpretable model that does not use the transformed input variables, a new clustering algorithm is proposed, based on the expectation-maximization (EM) identification of Gaussian mixture models. This new technique is applied to two well-known benchmark problems: the MPG (miles per gallon) prediction and a simulated second-order nonlinear process. The obtained results are compared with results from the literature.",
keywords = "Clustering methods, Expectation maximization, Fuzzy systems, Modeling, Optimization, Regression",
author = "J. Abonyi and Robert Babuška and F. Szeifert",
year = "2002",
month = "10",
doi = "10.1109/TSMCB.2002.1033180",
language = "English",
volume = "32",
pages = "612--621",
journal = "IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics",
issn = "1083-4419",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - Modified Gath-Geva fuzzy clustering for identification of Takagi-Sugeno fuzzy models

AU - Abonyi, J.

AU - Babuška, Robert

AU - Szeifert, F.

PY - 2002/10

Y1 - 2002/10

N2 - The construction of interpretable Takagi-Sugeno (TS) fuzzy models by means of clustering is addressed. First, it is shown how the antecedent fuzzy sets and the corresponding consequent parameters of the TS model can be derived from clusters obtained by the Gath-Geva (GG) algorithm. To preserve the partitioning of the antecedent space, linearly transformed input variables can be used in the model. This may, however, complicate the interpretation of the rules. To form an easily interpretable model that does not use the transformed input variables, a new clustering algorithm is proposed, based on the expectation-maximization (EM) identification of Gaussian mixture models. This new technique is applied to two well-known benchmark problems: the MPG (miles per gallon) prediction and a simulated second-order nonlinear process. The obtained results are compared with results from the literature.

AB - The construction of interpretable Takagi-Sugeno (TS) fuzzy models by means of clustering is addressed. First, it is shown how the antecedent fuzzy sets and the corresponding consequent parameters of the TS model can be derived from clusters obtained by the Gath-Geva (GG) algorithm. To preserve the partitioning of the antecedent space, linearly transformed input variables can be used in the model. This may, however, complicate the interpretation of the rules. To form an easily interpretable model that does not use the transformed input variables, a new clustering algorithm is proposed, based on the expectation-maximization (EM) identification of Gaussian mixture models. This new technique is applied to two well-known benchmark problems: the MPG (miles per gallon) prediction and a simulated second-order nonlinear process. The obtained results are compared with results from the literature.

KW - Clustering methods

KW - Expectation maximization

KW - Fuzzy systems

KW - Modeling

KW - Optimization

KW - Regression

UR - http://www.scopus.com/inward/record.url?scp=0036791593&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036791593&partnerID=8YFLogxK

U2 - 10.1109/TSMCB.2002.1033180

DO - 10.1109/TSMCB.2002.1033180

M3 - Article

C2 - 18244865

AN - SCOPUS:0036791593

VL - 32

SP - 612

EP - 621

JO - IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics

JF - IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics

SN - 1083-4419

IS - 5

ER -