K-means clustering seeds initialization based on centrality, sparsity, and isotropy

Pilsung Kang; Sungzoon Cho

doi:10.1007/978-3-642-04394-9_14

K-means clustering seeds initialization based on centrality, sparsity, and isotropy

Pilsung Kang, Sungzoon Cho

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

15 Citations (Scopus)

Abstract

K-Means is the most commonly used clustering algorithm. Despite its numerous advantages, it has a crucial drawback: the final cluster structure entirely relies on the choice of initial seeds. In this paper, a new seeds initialization algorithm based on centrality, sparsity, and isotropy is proposed. Preliminary experiments show that the proposed algorithm not only resulted in better clustering structures, but also accelerated the convergence.

Original language	English
Title of host publication	Intelligent Data Engineering and Automated Learning - IDEAL 2009 - 10th International Conference, Proceedings
Pages	109-117
Number of pages	9
DOIs	https://doi.org/10.1007/978-3-642-04394-9_14
Publication status	Published - 2009
Externally published	Yes
Event	10th International Conference on Intelligent Data Engineering and Automated Learning, IDEAL 2009 - Burgos, Spain Duration: 2009 Sept 23 → 2009 Sept 26

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	5788 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	10th International Conference on Intelligent Data Engineering and Automated Learning, IDEAL 2009
Country/Territory	Spain
City	Burgos
Period	09/9/23 → 09/9/26

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-642-04394-9_14

Cite this

Kang, P., & Cho, S. (2009). K-means clustering seeds initialization based on centrality, sparsity, and isotropy. In Intelligent Data Engineering and Automated Learning - IDEAL 2009 - 10th International Conference, Proceedings (pp. 109-117). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5788 LNCS). https://doi.org/10.1007/978-3-642-04394-9_14

K-means clustering seeds initialization based on centrality, sparsity, and isotropy. / Kang, Pilsung; Cho, Sungzoon.

Intelligent Data Engineering and Automated Learning - IDEAL 2009 - 10th International Conference, Proceedings. 2009. p. 109-117 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5788 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Kang, P & Cho, S 2009, K-means clustering seeds initialization based on centrality, sparsity, and isotropy. in Intelligent Data Engineering and Automated Learning - IDEAL 2009 - 10th International Conference, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5788 LNCS, pp. 109-117, 10th International Conference on Intelligent Data Engineering and Automated Learning, IDEAL 2009, Burgos, Spain, 09/9/23. https://doi.org/10.1007/978-3-642-04394-9_14

Kang P, Cho S. K-means clustering seeds initialization based on centrality, sparsity, and isotropy. In Intelligent Data Engineering and Automated Learning - IDEAL 2009 - 10th International Conference, Proceedings. 2009. p. 109-117. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-642-04394-9_14

Kang, Pilsung ; Cho, Sungzoon. / K-means clustering seeds initialization based on centrality, sparsity, and isotropy. Intelligent Data Engineering and Automated Learning - IDEAL 2009 - 10th International Conference, Proceedings. 2009. pp. 109-117 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{acca4ef5994e47b9a47c8a3373bdd980,

title = "K-means clustering seeds initialization based on centrality, sparsity, and isotropy",

abstract = "K-Means is the most commonly used clustering algorithm. Despite its numerous advantages, it has a crucial drawback: the final cluster structure entirely relies on the choice of initial seeds. In this paper, a new seeds initialization algorithm based on centrality, sparsity, and isotropy is proposed. Preliminary experiments show that the proposed algorithm not only resulted in better clustering structures, but also accelerated the convergence.",

author = "Pilsung Kang and Sungzoon Cho",

year = "2009",

doi = "10.1007/978-3-642-04394-9_14",

language = "English",

isbn = "3642043933",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "109--117",

booktitle = "Intelligent Data Engineering and Automated Learning - IDEAL 2009 - 10th International Conference, Proceedings",

note = "10th International Conference on Intelligent Data Engineering and Automated Learning, IDEAL 2009 ; Conference date: 23-09-2009 Through 26-09-2009",

}

TY - GEN

T1 - K-means clustering seeds initialization based on centrality, sparsity, and isotropy

AU - Kang, Pilsung

AU - Cho, Sungzoon

PY - 2009

Y1 - 2009

N2 - K-Means is the most commonly used clustering algorithm. Despite its numerous advantages, it has a crucial drawback: the final cluster structure entirely relies on the choice of initial seeds. In this paper, a new seeds initialization algorithm based on centrality, sparsity, and isotropy is proposed. Preliminary experiments show that the proposed algorithm not only resulted in better clustering structures, but also accelerated the convergence.

AB - K-Means is the most commonly used clustering algorithm. Despite its numerous advantages, it has a crucial drawback: the final cluster structure entirely relies on the choice of initial seeds. In this paper, a new seeds initialization algorithm based on centrality, sparsity, and isotropy is proposed. Preliminary experiments show that the proposed algorithm not only resulted in better clustering structures, but also accelerated the convergence.

UR - http://www.scopus.com/inward/record.url?scp=76249127819&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-04394-9_14

DO - 10.1007/978-3-642-04394-9_14

M3 - Conference contribution

AN - SCOPUS:76249127819

SN - 3642043933

SN - 9783642043932

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 109

EP - 117

BT - Intelligent Data Engineering and Automated Learning - IDEAL 2009 - 10th International Conference, Proceedings

T2 - 10th International Conference on Intelligent Data Engineering and Automated Learning, IDEAL 2009

Y2 - 23 September 2009 through 26 September 2009

ER -

K-means clustering seeds initialization based on centrality, sparsity, and isotropy

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this