On K-means algorithm with the use of Mahalanobis distances

Melnykov, Igor; Melnykov, Volodymyr

NUR Home
→
01.NU Schools
→
School of Science and Technology (2015-2019)
→
Mathematics
→
Articles
→
View Item

dc.contributor.author	Melnykov, Igor
dc.contributor.author	Melnykov, Volodymyr
dc.creator	Igor, Melnykov
dc.date.accessioned	2017-12-14T04:51:53Z
dc.date.available	2017-12-14T04:51:53Z
dc.date.issued	2014-01-01
dc.identifier	DOI:10.1016/j.spl.2013.09.026
dc.identifier.citation	Igor Melnykov, Volodymyr Melnykov, On K-means algorithm with the use of Mahalanobis distances, In Statistics & Probability Letters, Volume 84, 2014, Pages 88-95	en_US
dc.identifier.issn	01677152
dc.identifier.uri	https://www.sciencedirect.com/science/article/pii/S0167715213003246
dc.identifier.uri	http://nur.nu.edu.kz/handle/123456789/2887
dc.description.abstract	Abstract The K-means algorithm is commonly used with the Euclidean metric. While the use of Mahalanobis distances seems to be a straightforward extension of the algorithm, the initial estimation of covariance matrices can be complicated. We propose a novel approach for initializing covariance matrices.	en_US
dc.language.iso	en	en_US
dc.publisher	Statistics & Probability Letters	en_US
dc.relation.ispartof	Statistics & Probability Letters
dc.subject	K-means algorithm	en_US
dc.subject	Mahalanobis distance	en_US
dc.subject	Initialization	en_US
dc.title	On K-means algorithm with the use of Mahalanobis distances	en_US
dc.type	Article	en_US
dc.rights.license	Copyright © 2013 Elsevier B.V. All rights reserved.
elsevier.identifier.doi	10.1016/j.spl.2013.09.026
elsevier.identifier.eid	1-s2.0-S0167715213003246
elsevier.identifier.pii	S0167-7152(13)00324-6
elsevier.identifier.scopusid	84885983328
elsevier.volume	84
elsevier.coverdate	2014-01-01
elsevier.coverdisplaydate	January 2014
elsevier.startingpage	88
elsevier.endingpage	95
elsevier.openaccess	0
elsevier.openaccessarticle	false
elsevier.openarchivearticle	false
elsevier.teaser	The K-means algorithm is commonly used with the Euclidean metric. While the use of Mahalanobis distances seems to be a straightforward extension of the algorithm, the initial estimation of covariance...
elsevier.aggregationtype	Journal
workflow.import.source	science