A direct method for cluster analysis

S. S. Yau; S. C. Chang

doi:10.1016/0031-3203(75)90006-0

A direct method for cluster analysis

S. S. Yau, S. C. Chang

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

3 Scopus citations

Abstract

An algorithm based on a least-mean-square (LMS) criterion is presented. This algorithm partitions a multi-dimensional data set directly into a desired number of clusters. The result is compared favorably to existing methods in both performance and computational efficiency. An efficient method for determining a reasonable set of distributed initial cluster centers based on principal component analysis is also presented. This clustering algorithm is shown to converge to a unique minimum based on the LMS criterion and is demonstrated by digital computer simulation and applied to the analysis of vectorcardiograms.

Original language	English (US)
Pages (from-to)	215-224
Number of pages	10
Journal	Pattern Recognition
Volume	7
Issue number	4
DOIs	https://doi.org/10.1016/0031-3203(75)90006-0
State	Published - Dec 1975
Externally published	Yes

Keywords

Cluster analysis algorithm
Initial cluster centers
Least-mean-square criterion
Principal component analysis

ASJC Scopus subject areas

Software
Signal Processing
Computer Vision and Pattern Recognition
Artificial Intelligence

Access to Document

10.1016/0031-3203(75)90006-0

Cite this

@article{a79b822caa234c01b17dbfb8382a137c,

title = "A direct method for cluster analysis",

abstract = "An algorithm based on a least-mean-square (LMS) criterion is presented. This algorithm partitions a multi-dimensional data set directly into a desired number of clusters. The result is compared favorably to existing methods in both performance and computational efficiency. An efficient method for determining a reasonable set of distributed initial cluster centers based on principal component analysis is also presented. This clustering algorithm is shown to converge to a unique minimum based on the LMS criterion and is demonstrated by digital computer simulation and applied to the analysis of vectorcardiograms.",

keywords = "Cluster analysis algorithm, Initial cluster centers, Least-mean-square criterion, Principal component analysis",

author = "Yau, {S. S.} and Chang, {S. C.}",

year = "1975",

month = dec,

doi = "10.1016/0031-3203(75)90006-0",

language = "English (US)",

volume = "7",

pages = "215--224",

journal = "Pattern Recognition",

issn = "0031-3203",

publisher = "Elsevier Limited",

number = "4",

}

TY - JOUR

T1 - A direct method for cluster analysis

AU - Yau, S. S.

AU - Chang, S. C.

PY - 1975/12

Y1 - 1975/12

N2 - An algorithm based on a least-mean-square (LMS) criterion is presented. This algorithm partitions a multi-dimensional data set directly into a desired number of clusters. The result is compared favorably to existing methods in both performance and computational efficiency. An efficient method for determining a reasonable set of distributed initial cluster centers based on principal component analysis is also presented. This clustering algorithm is shown to converge to a unique minimum based on the LMS criterion and is demonstrated by digital computer simulation and applied to the analysis of vectorcardiograms.

AB - An algorithm based on a least-mean-square (LMS) criterion is presented. This algorithm partitions a multi-dimensional data set directly into a desired number of clusters. The result is compared favorably to existing methods in both performance and computational efficiency. An efficient method for determining a reasonable set of distributed initial cluster centers based on principal component analysis is also presented. This clustering algorithm is shown to converge to a unique minimum based on the LMS criterion and is demonstrated by digital computer simulation and applied to the analysis of vectorcardiograms.

KW - Cluster analysis algorithm

KW - Initial cluster centers

KW - Least-mean-square criterion

KW - Principal component analysis

UR - http://www.scopus.com/inward/record.url?scp=0016598641&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0016598641&partnerID=8YFLogxK

U2 - 10.1016/0031-3203(75)90006-0

DO - 10.1016/0031-3203(75)90006-0

M3 - Article

AN - SCOPUS:0016598641

SN - 0031-3203

VL - 7

SP - 215

EP - 224

JO - Pattern Recognition

JF - Pattern Recognition

IS - 4

ER -

A direct method for cluster analysis

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this