A new generalized learning vector quantization algorithm

Ching Tang Hsieh; Mu Chun Su; Uei Jyh Chen; Horng Jae Lee

A new generalized learning vector quantization algorithm

Ching Tang Hsieh, Mu Chun Su, Uei Jyh Chen, Horng Jae Lee

資訊工程學系

研究成果: 會議貢獻類型 › 會議論文 › 同行評審

1 引文斯高帕斯（Scopus）

摘要

A new approach of data clustering which is capable of detecting clusters of different shapes is proposed. In classical clustering approaches, it is a great challenge to separate clusters if the cluster prototypes are difficult to be represented by a mathematical formula. In this paper, we propose an improved learning vector quantization (LVQ) algorithm using the concept of symmetry. Through several computer simulations, the results show that the proposed method with the random initialization is effectiveness in detecting linear, spherical and ellipsoidal clusters. Besides, this method can also solve the crossed question.

原文	???core.languages.en_GB???
頁面	339-344
頁數	6
出版狀態	已出版 - 2000
事件	2000 IEEE Asia-Pacific Conference on Circuits and Systems: Electronic Communication Systems - Tianjin, China 持續時間: 4 12月 2000 → 6 12月 2000

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???	2000 IEEE Asia-Pacific Conference on Circuits and Systems: Electronic Communication Systems
國家/地區	China
城市	Tianjin
期間	4/12/00 → 6/12/00

引用此

@conference{2b43cfe38cda4f988b66dac96cdb7eac,

title = "A new generalized learning vector quantization algorithm",

abstract = "A new approach of data clustering which is capable of detecting clusters of different shapes is proposed. In classical clustering approaches, it is a great challenge to separate clusters if the cluster prototypes are difficult to be represented by a mathematical formula. In this paper, we propose an improved learning vector quantization (LVQ) algorithm using the concept of symmetry. Through several computer simulations, the results show that the proposed method with the random initialization is effectiveness in detecting linear, spherical and ellipsoidal clusters. Besides, this method can also solve the crossed question.",

author = "Hsieh, {Ching Tang} and Su, {Mu Chun} and Chen, {Uei Jyh} and Lee, {Horng Jae}",

year = "2000",

language = "???core.languages.en_GB???",

pages = "339--344",

note = "2000 IEEE Asia-Pacific Conference on Circuits and Systems: Electronic Communication Systems ; Conference date: 04-12-2000 Through 06-12-2000",

}

TY - CONF

T1 - A new generalized learning vector quantization algorithm

AU - Hsieh, Ching Tang

AU - Su, Mu Chun

AU - Chen, Uei Jyh

AU - Lee, Horng Jae

PY - 2000

Y1 - 2000

N2 - A new approach of data clustering which is capable of detecting clusters of different shapes is proposed. In classical clustering approaches, it is a great challenge to separate clusters if the cluster prototypes are difficult to be represented by a mathematical formula. In this paper, we propose an improved learning vector quantization (LVQ) algorithm using the concept of symmetry. Through several computer simulations, the results show that the proposed method with the random initialization is effectiveness in detecting linear, spherical and ellipsoidal clusters. Besides, this method can also solve the crossed question.

AB - A new approach of data clustering which is capable of detecting clusters of different shapes is proposed. In classical clustering approaches, it is a great challenge to separate clusters if the cluster prototypes are difficult to be represented by a mathematical formula. In this paper, we propose an improved learning vector quantization (LVQ) algorithm using the concept of symmetry. Through several computer simulations, the results show that the proposed method with the random initialization is effectiveness in detecting linear, spherical and ellipsoidal clusters. Besides, this method can also solve the crossed question.

UR - http://www.scopus.com/inward/record.url?scp=2342622686&partnerID=8YFLogxK

M3 - 會議論文

AN - SCOPUS:2342622686

SP - 339

EP - 344

T2 - 2000 IEEE Asia-Pacific Conference on Circuits and Systems: Electronic Communication Systems

Y2 - 4 December 2000 through 6 December 2000

ER -

A new generalized learning vector quantization algorithm

摘要

???event.eventtypes.event.conference???

指紋

引用此