Robust fuzzy clustering as a multi-objective optimization procedure

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

4 Citations (Scopus)

Abstract

In this paper, a multi-objective genetic algorithm for data clustering based on the robust fuzzy least trimmed squares estimator is proposed. The clustering methodology addresses two critical issues in unsupervised data clustering - the ability to produce meaningful classification in noisy data, and the requirement that the number of clusters be known a priori. The GA-driven clustering routine optimizes number of clusters as well as cluster assignment, and cluster prototypes. A twoparameter, mapped, fixed point coding scheme is used to represent assignment of data into either the true retained set and the noisy trimmed set, and the optimal number of clusters in the retained set. A three-objective criterion is used as the minimization functional for the GA. Results on well-known data sets from literature suggest that the proposed methodology is comparable (in many cases superior) to conventional robust fuzzy clustering algorithms that assume a known value for optimal number of clusters.

Original languageEnglish (US)
Title of host publicationNAFIPS 2009 - 2009 Annual Meeting of the North American Fuzzy Information Processing Society
DOIs
StatePublished - Nov 2 2009
Event2009 Annual Meeting of the North American Fuzzy Information Processing Society, NAFIPS 2009 - Cincinnati, OH, United States
Duration: Jun 14 2009Jun 17 2009

Other

Other2009 Annual Meeting of the North American Fuzzy Information Processing Society, NAFIPS 2009
CountryUnited States
CityCincinnati, OH
Period6/14/096/17/09

Fingerprint

Fuzzy clustering
Fuzzy Clustering
Number of Clusters
Multiobjective optimization
Multi-objective Optimization
Clustering algorithms
Genetic algorithms
Data Clustering
Assignment
Least Trimmed Squares
Clustering
Unsupervised Clustering
Multi-objective Genetic Algorithm
Methodology
Fuzzy Algorithm
Noisy Data
Clustering Algorithm
Coding
Fixed point
Optimise

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Mathematics(all)

Cite this

Banerjee, A. (2009). Robust fuzzy clustering as a multi-objective optimization procedure. In NAFIPS 2009 - 2009 Annual Meeting of the North American Fuzzy Information Processing Society [5156399] https://doi.org/10.1109/NAFIPS.2009.5156399
Banerjee, Amit. / Robust fuzzy clustering as a multi-objective optimization procedure. NAFIPS 2009 - 2009 Annual Meeting of the North American Fuzzy Information Processing Society. 2009.
@inproceedings{3c1fd2028aa5418da7be934cf6afa33b,
title = "Robust fuzzy clustering as a multi-objective optimization procedure",
abstract = "In this paper, a multi-objective genetic algorithm for data clustering based on the robust fuzzy least trimmed squares estimator is proposed. The clustering methodology addresses two critical issues in unsupervised data clustering - the ability to produce meaningful classification in noisy data, and the requirement that the number of clusters be known a priori. The GA-driven clustering routine optimizes number of clusters as well as cluster assignment, and cluster prototypes. A twoparameter, mapped, fixed point coding scheme is used to represent assignment of data into either the true retained set and the noisy trimmed set, and the optimal number of clusters in the retained set. A three-objective criterion is used as the minimization functional for the GA. Results on well-known data sets from literature suggest that the proposed methodology is comparable (in many cases superior) to conventional robust fuzzy clustering algorithms that assume a known value for optimal number of clusters.",
author = "Amit Banerjee",
year = "2009",
month = "11",
day = "2",
doi = "10.1109/NAFIPS.2009.5156399",
language = "English (US)",
isbn = "9781424445776",
booktitle = "NAFIPS 2009 - 2009 Annual Meeting of the North American Fuzzy Information Processing Society",

}

Banerjee, A 2009, Robust fuzzy clustering as a multi-objective optimization procedure. in NAFIPS 2009 - 2009 Annual Meeting of the North American Fuzzy Information Processing Society., 5156399, 2009 Annual Meeting of the North American Fuzzy Information Processing Society, NAFIPS 2009, Cincinnati, OH, United States, 6/14/09. https://doi.org/10.1109/NAFIPS.2009.5156399

Robust fuzzy clustering as a multi-objective optimization procedure. / Banerjee, Amit.

NAFIPS 2009 - 2009 Annual Meeting of the North American Fuzzy Information Processing Society. 2009. 5156399.

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

TY - GEN

T1 - Robust fuzzy clustering as a multi-objective optimization procedure

AU - Banerjee, Amit

PY - 2009/11/2

Y1 - 2009/11/2

N2 - In this paper, a multi-objective genetic algorithm for data clustering based on the robust fuzzy least trimmed squares estimator is proposed. The clustering methodology addresses two critical issues in unsupervised data clustering - the ability to produce meaningful classification in noisy data, and the requirement that the number of clusters be known a priori. The GA-driven clustering routine optimizes number of clusters as well as cluster assignment, and cluster prototypes. A twoparameter, mapped, fixed point coding scheme is used to represent assignment of data into either the true retained set and the noisy trimmed set, and the optimal number of clusters in the retained set. A three-objective criterion is used as the minimization functional for the GA. Results on well-known data sets from literature suggest that the proposed methodology is comparable (in many cases superior) to conventional robust fuzzy clustering algorithms that assume a known value for optimal number of clusters.

AB - In this paper, a multi-objective genetic algorithm for data clustering based on the robust fuzzy least trimmed squares estimator is proposed. The clustering methodology addresses two critical issues in unsupervised data clustering - the ability to produce meaningful classification in noisy data, and the requirement that the number of clusters be known a priori. The GA-driven clustering routine optimizes number of clusters as well as cluster assignment, and cluster prototypes. A twoparameter, mapped, fixed point coding scheme is used to represent assignment of data into either the true retained set and the noisy trimmed set, and the optimal number of clusters in the retained set. A three-objective criterion is used as the minimization functional for the GA. Results on well-known data sets from literature suggest that the proposed methodology is comparable (in many cases superior) to conventional robust fuzzy clustering algorithms that assume a known value for optimal number of clusters.

UR - http://www.scopus.com/inward/record.url?scp=70350424370&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70350424370&partnerID=8YFLogxK

U2 - 10.1109/NAFIPS.2009.5156399

DO - 10.1109/NAFIPS.2009.5156399

M3 - Conference contribution

SN - 9781424445776

BT - NAFIPS 2009 - 2009 Annual Meeting of the North American Fuzzy Information Processing Society

ER -

Banerjee A. Robust fuzzy clustering as a multi-objective optimization procedure. In NAFIPS 2009 - 2009 Annual Meeting of the North American Fuzzy Information Processing Society. 2009. 5156399 https://doi.org/10.1109/NAFIPS.2009.5156399