K*-Means: A new generalized k-means clustering algorithm

Yiu Ming CHEUNG

doi:10.1016/S0167-8655(03)00146-6

K*-Means: A new generalized k-means clustering algorithm

Yiu Ming CHEUNG^*

^*Corresponding author for this work

Department of Computer Science

Research output: Contribution to journal › Journal article › peer-review

166 Citations (Scopus)

Abstract

This paper presents a generalized version of the conventional k-means clustering algorithm [Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability, 1, University of California Press, Berkeley, 1967, p. 281]. Not only is this new one applicable to ellipse-shaped data clusters without dead-unit problem, but also performs correct clustering without pre-assigning the exact cluster number. We qualitatively analyze its underlying mechanism, and show its outstanding performance through the experiments.

Original language	English
Pages (from-to)	2883-2893
Number of pages	11
Journal	Pattern Recognition Letters
Volume	24
Issue number	15
DOIs	https://doi.org/10.1016/S0167-8655(03)00146-6
Publication status	Published - Nov 2003

Scopus Subject Areas

Software
Signal Processing
Computer Vision and Pattern Recognition
Artificial Intelligence

User-Defined Keywords

Cluster number
Clustering analysis
K-Means algorithm
Rival penalization

Access to Document

10.1016/S0167-8655(03)00146-6

Cite this

@article{c02fc4099bef4bd0bf5075f9bf96dcb1,

title = "K*-Means: A new generalized k-means clustering algorithm",

abstract = "This paper presents a generalized version of the conventional k-means clustering algorithm [Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability, 1, University of California Press, Berkeley, 1967, p. 281]. Not only is this new one applicable to ellipse-shaped data clusters without dead-unit problem, but also performs correct clustering without pre-assigning the exact cluster number. We qualitatively analyze its underlying mechanism, and show its outstanding performance through the experiments.",

keywords = "Cluster number, Clustering analysis, K-Means algorithm, Rival penalization",

author = "CHEUNG, {Yiu Ming}",

note = "Funding Information: This work was supported by a Faculty Research Grant of Hong Kong Baptist University with the project code: FRG/02-03/I-06. ",

year = "2003",

month = nov,

doi = "10.1016/S0167-8655(03)00146-6",

language = "English",

volume = "24",

pages = "2883--2893",

journal = "Pattern Recognition Letters",

issn = "0167-8655",

publisher = "Elsevier",

number = "15",

}

TY - JOUR

T1 - K*-Means

T2 - A new generalized k-means clustering algorithm

AU - CHEUNG, Yiu Ming

N1 - Funding Information: This work was supported by a Faculty Research Grant of Hong Kong Baptist University with the project code: FRG/02-03/I-06.

PY - 2003/11

Y1 - 2003/11

N2 - This paper presents a generalized version of the conventional k-means clustering algorithm [Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability, 1, University of California Press, Berkeley, 1967, p. 281]. Not only is this new one applicable to ellipse-shaped data clusters without dead-unit problem, but also performs correct clustering without pre-assigning the exact cluster number. We qualitatively analyze its underlying mechanism, and show its outstanding performance through the experiments.

AB - This paper presents a generalized version of the conventional k-means clustering algorithm [Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability, 1, University of California Press, Berkeley, 1967, p. 281]. Not only is this new one applicable to ellipse-shaped data clusters without dead-unit problem, but also performs correct clustering without pre-assigning the exact cluster number. We qualitatively analyze its underlying mechanism, and show its outstanding performance through the experiments.

KW - Cluster number

KW - Clustering analysis

KW - K-Means algorithm

KW - Rival penalization

UR - http://www.scopus.com/inward/record.url?scp=0041328214&partnerID=8YFLogxK

U2 - 10.1016/S0167-8655(03)00146-6

DO - 10.1016/S0167-8655(03)00146-6

M3 - Journal article

AN - SCOPUS:0041328214

SN - 0167-8655

VL - 24

SP - 2883

EP - 2893

JO - Pattern Recognition Letters

JF - Pattern Recognition Letters

IS - 15

ER -

K*-Means: A new generalized k-means clustering algorithm

Abstract

Scopus Subject Areas

User-Defined Keywords

Access to Document

Other files and links

Fingerprint

Cite this