An Empirical Evaluation of Different Initializations on the Number of K-Means Iterations

Renato Cordeiro De Amorim

doi:10.1007/978-3-642-37807-2_2

An Empirical Evaluation of Different Initializations on the Number of K-Means Iterations

Renato Cordeiro De Amorim

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

6 Citations (Scopus)

Abstract

This paper presents an analysis of the number of iterations K-Means takes to converge under different initializations. We have experimented with seven initialization algorithms in a total of 37 real and synthetic datasets. We have found that hierarchical-based initializations tend to be most effective at reducing the number of iterations, especially a divisive algorithm using the Ward criterion when applied to real datasets

Original language	English
Title of host publication	Advances in Artificial Intelligence
Subtitle of host publication	11th Mexican International Conference on Artificial Intelligence - revised selected papers
Publisher	Springer Nature
Pages	15-26
ISBN (Electronic)	978-3-642-37807-2
ISBN (Print)	978-3-642-37806-5
DOIs	https://doi.org/10.1007/978-3-642-37807-2_2
Publication status	Published - 2013
Event	11th Mexican Int Conf on Artificial Intelligence - MICAI 2012 - San Luis Potosi, Mexico Duration: 27 Oct 2012 → 4 Nov 2012

Publication series

Name	Lecture Notes in Computer Science
Volume	7629

Conference

Conference	11th Mexican Int Conf on Artificial Intelligence - MICAI 2012
Country/Territory	Mexico
City	San Luis Potosi
Period	27/10/12 → 4/11/12

Access to Document

10.1007/978-3-642-37807-2_2

Cite this

Cordeiro De Amorim, R. (2013). An Empirical Evaluation of Different Initializations on the Number of K-Means Iterations. In Advances in Artificial Intelligence: 11th Mexican International Conference on Artificial Intelligence - revised selected papers (pp. 15-26). (Lecture Notes in Computer Science; Vol. 7629). Springer Nature. https://doi.org/10.1007/978-3-642-37807-2_2

@inproceedings{d5482480ae55401c950e21b879196797,

title = "An Empirical Evaluation of Different Initializations on the Number of K-Means Iterations",

abstract = "This paper presents an analysis of the number of iterations K-Means takes to converge under different initializations. We have experimented with seven initialization algorithms in a total of 37 real and synthetic datasets. We have found that hierarchical-based initializations tend to be most effective at reducing the number of iterations, especially a divisive algorithm using the Ward criterion when applied to real datasets",

author = "{Cordeiro De Amorim}, Renato",

year = "2013",

doi = "10.1007/978-3-642-37807-2_2",

language = "English",

isbn = "978-3-642-37806-5",

series = "Lecture Notes in Computer Science",

publisher = "Springer Nature",

pages = "15--26",

booktitle = "Advances in Artificial Intelligence",

address = "Netherlands",

note = "11th Mexican Int Conf on Artificial Intelligence - MICAI 2012 ; Conference date: 27-10-2012 Through 04-11-2012",

}

Cordeiro De Amorim, R 2013, An Empirical Evaluation of Different Initializations on the Number of K-Means Iterations. in Advances in Artificial Intelligence: 11th Mexican International Conference on Artificial Intelligence - revised selected papers. Lecture Notes in Computer Science, vol. 7629, Springer Nature, pp. 15-26, 11th Mexican Int Conf on Artificial Intelligence - MICAI 2012, San Luis Potosi, Mexico, 27/10/12. https://doi.org/10.1007/978-3-642-37807-2_2

An Empirical Evaluation of Different Initializations on the Number of K-Means Iterations. / Cordeiro De Amorim, Renato.
Advances in Artificial Intelligence: 11th Mexican International Conference on Artificial Intelligence - revised selected papers. Springer Nature, 2013. p. 15-26 (Lecture Notes in Computer Science; Vol. 7629).

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

TY - GEN

T1 - An Empirical Evaluation of Different Initializations on the Number of K-Means Iterations

AU - Cordeiro De Amorim, Renato

PY - 2013

Y1 - 2013

N2 - This paper presents an analysis of the number of iterations K-Means takes to converge under different initializations. We have experimented with seven initialization algorithms in a total of 37 real and synthetic datasets. We have found that hierarchical-based initializations tend to be most effective at reducing the number of iterations, especially a divisive algorithm using the Ward criterion when applied to real datasets

AB - This paper presents an analysis of the number of iterations K-Means takes to converge under different initializations. We have experimented with seven initialization algorithms in a total of 37 real and synthetic datasets. We have found that hierarchical-based initializations tend to be most effective at reducing the number of iterations, especially a divisive algorithm using the Ward criterion when applied to real datasets

U2 - 10.1007/978-3-642-37807-2_2

DO - 10.1007/978-3-642-37807-2_2

M3 - Conference contribution

SN - 978-3-642-37806-5

T3 - Lecture Notes in Computer Science

SP - 15

EP - 26

BT - Advances in Artificial Intelligence

PB - Springer Nature

T2 - 11th Mexican Int Conf on Artificial Intelligence - MICAI 2012

Y2 - 27 October 2012 through 4 November 2012

ER -