Double partition around medoids based cluster ensemble

Le Li; Jia You; Guoqiang Han; Hantao Chen

doi:10.1109/ICMLC.2012.6359568

Double partition around medoids based cluster ensemble

Le Li, Jia You, Guoqiang Han, Hantao Chen

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

3 Citations (Scopus)

Abstract

Cluster ensemble is one of the hot topics in the machine learning area. Though plenty of cluster ensemble methods and frameworks have been proposed, many cluster ensemble methods are easily faded by noisy datasets and local optimal problems. In this article, we introduced a novel cluster ensemble method, named as Double Partition Around Medoids based Cluster Ensemble (PAM2CE). PAM2CE will effectively weaken or even eliminate the effect of noisy datasets and local optimal problems via clustering attributes and selecting the representative attributes. The experimental results reveal the better robustness and effectiveness of proposed method.

Original language	English
Title of host publication	Proceedings of 2012 International Conference on Machine Learning and Cybernetics, ICMLC 2012
Pages	1390-1394
Number of pages	5
Volume	4
DOIs	https://doi.org/10.1109/ICMLC.2012.6359568
Publication status	Published - 31 Dec 2012
Event	2012 International Conference on Machine Learning and Cybernetics, ICMLC 2012 - Xian, Shaanxi, China Duration: 15 Jul 2012 → 17 Jul 2012

Conference

Conference	2012 International Conference on Machine Learning and Cybernetics, ICMLC 2012
Country/Territory	China
City	Xian, Shaanxi
Period	15/07/12 → 17/07/12

ASJC Scopus subject areas

Artificial Intelligence
Computational Theory and Mathematics
Computer Networks and Communications
Human-Computer Interaction

More information

10.1109/ICMLC.2012.6359568

Cite this

@inproceedings{29af50f8ab3b4c80a7c15907250426c3,

title = "Double partition around medoids based cluster ensemble",

abstract = "Cluster ensemble is one of the hot topics in the machine learning area. Though plenty of cluster ensemble methods and frameworks have been proposed, many cluster ensemble methods are easily faded by noisy datasets and local optimal problems. In this article, we introduced a novel cluster ensemble method, named as Double Partition Around Medoids based Cluster Ensemble (PAM2CE). PAM2CE will effectively weaken or even eliminate the effect of noisy datasets and local optimal problems via clustering attributes and selecting the representative attributes. The experimental results reveal the better robustness and effectiveness of proposed method.",

author = "Le Li and Jia You and Guoqiang Han and Hantao Chen",

year = "2012",

month = dec,

day = "31",

doi = "10.1109/ICMLC.2012.6359568",

language = "English",

isbn = "9781467314855",

volume = "4",

pages = "1390--1394",

booktitle = "Proceedings of 2012 International Conference on Machine Learning and Cybernetics, ICMLC 2012",

note = "2012 International Conference on Machine Learning and Cybernetics, ICMLC 2012 ; Conference date: 15-07-2012 Through 17-07-2012",

}

Li, L, You, J, Han, G & Chen, H 2012, Double partition around medoids based cluster ensemble. in Proceedings of 2012 International Conference on Machine Learning and Cybernetics, ICMLC 2012. vol. 4, 6359568, pp. 1390-1394, 2012 International Conference on Machine Learning and Cybernetics, ICMLC 2012, Xian, Shaanxi, China, 15/07/12. https://doi.org/10.1109/ICMLC.2012.6359568

Double partition around medoids based cluster ensemble. / Li, Le; You, Jia; Han, Guoqiang et al.
Proceedings of 2012 International Conference on Machine Learning and Cybernetics, ICMLC 2012. Vol. 4 2012. p. 1390-1394 6359568.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Double partition around medoids based cluster ensemble

AU - Li, Le

AU - You, Jia

AU - Han, Guoqiang

AU - Chen, Hantao

PY - 2012/12/31

Y1 - 2012/12/31

N2 - Cluster ensemble is one of the hot topics in the machine learning area. Though plenty of cluster ensemble methods and frameworks have been proposed, many cluster ensemble methods are easily faded by noisy datasets and local optimal problems. In this article, we introduced a novel cluster ensemble method, named as Double Partition Around Medoids based Cluster Ensemble (PAM2CE). PAM2CE will effectively weaken or even eliminate the effect of noisy datasets and local optimal problems via clustering attributes and selecting the representative attributes. The experimental results reveal the better robustness and effectiveness of proposed method.

AB - Cluster ensemble is one of the hot topics in the machine learning area. Though plenty of cluster ensemble methods and frameworks have been proposed, many cluster ensemble methods are easily faded by noisy datasets and local optimal problems. In this article, we introduced a novel cluster ensemble method, named as Double Partition Around Medoids based Cluster Ensemble (PAM2CE). PAM2CE will effectively weaken or even eliminate the effect of noisy datasets and local optimal problems via clustering attributes and selecting the representative attributes. The experimental results reveal the better robustness and effectiveness of proposed method.

UR - http://www.scopus.com/inward/record.url?scp=84871582486&partnerID=8YFLogxK

U2 - 10.1109/ICMLC.2012.6359568

DO - 10.1109/ICMLC.2012.6359568

M3 - Conference article published in proceeding or book

SN - 9781467314855

VL - 4

SP - 1390

EP - 1394

BT - Proceedings of 2012 International Conference on Machine Learning and Cybernetics, ICMLC 2012

T2 - 2012 International Conference on Machine Learning and Cybernetics, ICMLC 2012

Y2 - 15 July 2012 through 17 July 2012

ER -

Double partition around medoids based cluster ensemble

Abstract

Conference

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this