Clustering by genetic algorithm: High quality chromosome selection for initial population

Research output: Book chapter/Published conference paperConference paper

6 Citations (Scopus)

Abstract

Many existing GA based clustering techniques generate the number of genes for a chromosome through random selection. Due the random selection there is a high chance of getting poor quality of initial genes in the initial population. A poor-quality initial population is likely to produce a poor-quality clustering solution. We argue that having a set of high quality chromosomes in the initial population we are more likely to produce a clustering solution of higher quality. Therefore, in this paper we propose a genetic algorithm based clustering technique that produces high quality initial chromosomes. The proposed technique selects the first 50% of the chromosomes through a deterministic selection phase and the remaining 50% chromosomes through a random selection phase, for the initial population. The proposed technique also uses crossover and mutation operation to getting better clustering result. We conduct experiments on seven datasets that are available in UCI machine learning repository. Two evaluation criteria namely silhouette coefficient and DB index are used. Our experiment results, based on the two evaluation criteria indicate a clear superiority of our technique over three existing techniques namely AGCUK, GAGR and K-means.
Original languageEnglish
Title of host publicationProceedings of the 2015 IEEE 10th Conference on Industrial Electronics and Applications (ICIEA)
Place of PublicationUnited States
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages129-134
Number of pages6
DOIs
Publication statusPublished - 2015
EventIEEE Conference on Industrial Electronics and Applications - Crowne Plaza, Auckland, New Zealand
Duration: 15 Jun 201517 Jun 2015
http://www.ieeeiciea.org/2015/

Conference

ConferenceIEEE Conference on Industrial Electronics and Applications
CountryNew Zealand
CityAuckland
Period15/06/1517/06/15
Internet address

Fingerprint

Chromosomes
Genetic algorithms
Genes
Learning systems
Experiments

Cite this

Beg, A. H., & Islam, M. Z. (2015). Clustering by genetic algorithm: High quality chromosome selection for initial population. In Proceedings of the 2015 IEEE 10th Conference on Industrial Electronics and Applications (ICIEA) (pp. 129-134). United States: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICIEA.2015.7334097
Beg, Abul Hashem ; Islam, Md Zahidul. / Clustering by genetic algorithm : High quality chromosome selection for initial population. Proceedings of the 2015 IEEE 10th Conference on Industrial Electronics and Applications (ICIEA). United States : IEEE, Institute of Electrical and Electronics Engineers, 2015. pp. 129-134
@inproceedings{e82f7b9faa4d47ceba0a3a170a58d905,
title = "Clustering by genetic algorithm: High quality chromosome selection for initial population",
abstract = "Many existing GA based clustering techniques generate the number of genes for a chromosome through random selection. Due the random selection there is a high chance of getting poor quality of initial genes in the initial population. A poor-quality initial population is likely to produce a poor-quality clustering solution. We argue that having a set of high quality chromosomes in the initial population we are more likely to produce a clustering solution of higher quality. Therefore, in this paper we propose a genetic algorithm based clustering technique that produces high quality initial chromosomes. The proposed technique selects the first 50{\%} of the chromosomes through a deterministic selection phase and the remaining 50{\%} chromosomes through a random selection phase, for the initial population. The proposed technique also uses crossover and mutation operation to getting better clustering result. We conduct experiments on seven datasets that are available in UCI machine learning repository. Two evaluation criteria namely silhouette coefficient and DB index are used. Our experiment results, based on the two evaluation criteria indicate a clear superiority of our technique over three existing techniques namely AGCUK, GAGR and K-means.",
author = "Beg, {Abul Hashem} and Islam, {Md Zahidul}",
note = "Imported on 03 May 2017 - DigiTool details were: publisher = 2015. Event dates (773o) = 15-06-2015-17-06-2015; Parent title (773t) = IEEE Conference on Industrial Electronics and Applications.",
year = "2015",
doi = "10.1109/ICIEA.2015.7334097",
language = "English",
pages = "129--134",
booktitle = "Proceedings of the 2015 IEEE 10th Conference on Industrial Electronics and Applications (ICIEA)",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States",

}

Beg, AH & Islam, MZ 2015, Clustering by genetic algorithm: High quality chromosome selection for initial population. in Proceedings of the 2015 IEEE 10th Conference on Industrial Electronics and Applications (ICIEA). IEEE, Institute of Electrical and Electronics Engineers, United States, pp. 129-134, IEEE Conference on Industrial Electronics and Applications, Auckland, New Zealand, 15/06/15. https://doi.org/10.1109/ICIEA.2015.7334097

Clustering by genetic algorithm : High quality chromosome selection for initial population. / Beg, Abul Hashem; Islam, Md Zahidul.

Proceedings of the 2015 IEEE 10th Conference on Industrial Electronics and Applications (ICIEA). United States : IEEE, Institute of Electrical and Electronics Engineers, 2015. p. 129-134.

Research output: Book chapter/Published conference paperConference paper

TY - GEN

T1 - Clustering by genetic algorithm

T2 - High quality chromosome selection for initial population

AU - Beg, Abul Hashem

AU - Islam, Md Zahidul

N1 - Imported on 03 May 2017 - DigiTool details were: publisher = 2015. Event dates (773o) = 15-06-2015-17-06-2015; Parent title (773t) = IEEE Conference on Industrial Electronics and Applications.

PY - 2015

Y1 - 2015

N2 - Many existing GA based clustering techniques generate the number of genes for a chromosome through random selection. Due the random selection there is a high chance of getting poor quality of initial genes in the initial population. A poor-quality initial population is likely to produce a poor-quality clustering solution. We argue that having a set of high quality chromosomes in the initial population we are more likely to produce a clustering solution of higher quality. Therefore, in this paper we propose a genetic algorithm based clustering technique that produces high quality initial chromosomes. The proposed technique selects the first 50% of the chromosomes through a deterministic selection phase and the remaining 50% chromosomes through a random selection phase, for the initial population. The proposed technique also uses crossover and mutation operation to getting better clustering result. We conduct experiments on seven datasets that are available in UCI machine learning repository. Two evaluation criteria namely silhouette coefficient and DB index are used. Our experiment results, based on the two evaluation criteria indicate a clear superiority of our technique over three existing techniques namely AGCUK, GAGR and K-means.

AB - Many existing GA based clustering techniques generate the number of genes for a chromosome through random selection. Due the random selection there is a high chance of getting poor quality of initial genes in the initial population. A poor-quality initial population is likely to produce a poor-quality clustering solution. We argue that having a set of high quality chromosomes in the initial population we are more likely to produce a clustering solution of higher quality. Therefore, in this paper we propose a genetic algorithm based clustering technique that produces high quality initial chromosomes. The proposed technique selects the first 50% of the chromosomes through a deterministic selection phase and the remaining 50% chromosomes through a random selection phase, for the initial population. The proposed technique also uses crossover and mutation operation to getting better clustering result. We conduct experiments on seven datasets that are available in UCI machine learning repository. Two evaluation criteria namely silhouette coefficient and DB index are used. Our experiment results, based on the two evaluation criteria indicate a clear superiority of our technique over three existing techniques namely AGCUK, GAGR and K-means.

U2 - 10.1109/ICIEA.2015.7334097

DO - 10.1109/ICIEA.2015.7334097

M3 - Conference paper

SP - 129

EP - 134

BT - Proceedings of the 2015 IEEE 10th Conference on Industrial Electronics and Applications (ICIEA)

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - United States

ER -

Beg AH, Islam MZ. Clustering by genetic algorithm: High quality chromosome selection for initial population. In Proceedings of the 2015 IEEE 10th Conference on Industrial Electronics and Applications (ICIEA). United States: IEEE, Institute of Electrical and Electronics Engineers. 2015. p. 129-134 https://doi.org/10.1109/ICIEA.2015.7334097