CLAM: Clustering large applications using metaheuristics

Quynh Nguyen; V. J. Rayward-Smith

doi:10.1007/s10852-010-9141-1

CLAM: Clustering large applications using metaheuristics

Quynh Nguyen, V. J. Rayward-Smith

School of Computing Sciences

Research output: Contribution to journal › Article › peer-review

5 Citations (Scopus)

Abstract

Clustering remains one of the most difficult challenges in data mining. This paper proposes a new algorithm, CLAM, using a hybrid metaheuristic between VNS and Tabu Search to solve the problem of k-medoid clustering. The new technique is compared to the well-known CLARANS. Experimental results show that, given the same computation times, CLAM is more effective.

Original language	English
Pages (from-to)	57-78
Number of pages	22
Journal	Journal of Mathematical Modelling and Algorithms
Volume	10
Issue number	1
DOIs	https://doi.org/10.1007/s10852-010-9141-1
Publication status	Published - Mar 2011

Access to Document

10.1007/s10852-010-9141-1

Cite this

@article{b86c4fa21e80483d8a0ab08987d50a34,

title = "CLAM: Clustering large applications using metaheuristics",

abstract = "Clustering remains one of the most difficult challenges in data mining. This paper proposes a new algorithm, CLAM, using a hybrid metaheuristic between VNS and Tabu Search to solve the problem of k-medoid clustering. The new technique is compared to the well-known CLARANS. Experimental results show that, given the same computation times, CLAM is more effective.",

author = "Quynh Nguyen and Rayward-Smith, {V. J.}",

year = "2011",

month = mar,

doi = "10.1007/s10852-010-9141-1",

language = "English",

volume = "10",

pages = "57--78",

journal = "Journal of Mathematical Modelling and Algorithms",

issn = "1570-1166",

publisher = "Springer",

number = "1",

}

TY - JOUR

T1 - CLAM: Clustering large applications using metaheuristics

AU - Nguyen, Quynh

AU - Rayward-Smith, V. J.

PY - 2011/3

Y1 - 2011/3

N2 - Clustering remains one of the most difficult challenges in data mining. This paper proposes a new algorithm, CLAM, using a hybrid metaheuristic between VNS and Tabu Search to solve the problem of k-medoid clustering. The new technique is compared to the well-known CLARANS. Experimental results show that, given the same computation times, CLAM is more effective.

AB - Clustering remains one of the most difficult challenges in data mining. This paper proposes a new algorithm, CLAM, using a hybrid metaheuristic between VNS and Tabu Search to solve the problem of k-medoid clustering. The new technique is compared to the well-known CLARANS. Experimental results show that, given the same computation times, CLAM is more effective.

U2 - 10.1007/s10852-010-9141-1

DO - 10.1007/s10852-010-9141-1

M3 - Article

SN - 1570-1166

VL - 10

SP - 57

EP - 78

JO - Journal of Mathematical Modelling and Algorithms

JF - Journal of Mathematical Modelling and Algorithms

IS - 1

ER -