A non-parametric maximum entropy clustering

Hideitsu Hino; Noboru Murata

doi:10.1007/978-3-319-11179-7_15

A non-parametric maximum entropy clustering

Hideitsu Hino, Noboru Murata

School of Advanced Science and Engineering

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

Abstract

Clustering is a fundamental tool for exploratory data analysis. Information theoretic clustering is based on the optimization of information theoretic quantities such as entropy and mutual information. Recently, since these quantities can be estimated in non-parametric manner, non-parametric information theoretic clustering gains much attention. Assuming the dataset is sampled from a certain cluster, and assigning different sampling weights depending on the clusters, the cluster conditional information theoretic quantities are estimated. In this paper, a simple clustering algorithm is proposed based on the principle of maximum entropy. The algorithm is experimentally shown to be comparable to or outperform conventional non-parametric clustering methods.

Original language	English
Title of host publication	Artificial Neural Networks and Machine Learning, ICANN 2014 - 24th International Conference on Artificial Neural Networks, Proceedings
Publisher	Springer Verlag
Pages	113-120
Number of pages	8
ISBN (Print)	9783319111780
DOIs	https://doi.org/10.1007/978-3-319-11179-7_15
Publication status	Published - 2014
Event	24th International Conference on Artificial Neural Networks, ICANN 2014 - Hamburg, Germany Duration: 2014 Sept 15 → 2014 Sept 19

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	8681 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	24th International Conference on Artificial Neural Networks, ICANN 2014
Country/Territory	Germany
City	Hamburg
Period	14/9/15 → 14/9/19

Keywords

Information Theoretic Clustering
Likelihood and Entropy estimator
Non-parametric

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-319-11179-7_15

Cite this

Hino, H., & Murata, N. (2014). A non-parametric maximum entropy clustering. In Artificial Neural Networks and Machine Learning, ICANN 2014 - 24th International Conference on Artificial Neural Networks, Proceedings (pp. 113-120). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8681 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-11179-7_15

A non-parametric maximum entropy clustering. / Hino, Hideitsu; Murata, Noboru.
Artificial Neural Networks and Machine Learning, ICANN 2014 - 24th International Conference on Artificial Neural Networks, Proceedings. Springer Verlag, 2014. p. 113-120 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8681 LNCS).

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

Hino, H & Murata, N 2014, A non-parametric maximum entropy clustering. in Artificial Neural Networks and Machine Learning, ICANN 2014 - 24th International Conference on Artificial Neural Networks, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8681 LNCS, Springer Verlag, pp. 113-120, 24th International Conference on Artificial Neural Networks, ICANN 2014, Hamburg, Germany, 14/9/15. https://doi.org/10.1007/978-3-319-11179-7_15

Hino H, Murata N. A non-parametric maximum entropy clustering. In Artificial Neural Networks and Machine Learning, ICANN 2014 - 24th International Conference on Artificial Neural Networks, Proceedings. Springer Verlag. 2014. p. 113-120. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-11179-7_15

Hino, Hideitsu ; Murata, Noboru. / A non-parametric maximum entropy clustering. Artificial Neural Networks and Machine Learning, ICANN 2014 - 24th International Conference on Artificial Neural Networks, Proceedings. Springer Verlag, 2014. pp. 113-120 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{c859021a29ca449891136aadc92ded9c,

title = "A non-parametric maximum entropy clustering",

abstract = "Clustering is a fundamental tool for exploratory data analysis. Information theoretic clustering is based on the optimization of information theoretic quantities such as entropy and mutual information. Recently, since these quantities can be estimated in non-parametric manner, non-parametric information theoretic clustering gains much attention. Assuming the dataset is sampled from a certain cluster, and assigning different sampling weights depending on the clusters, the cluster conditional information theoretic quantities are estimated. In this paper, a simple clustering algorithm is proposed based on the principle of maximum entropy. The algorithm is experimentally shown to be comparable to or outperform conventional non-parametric clustering methods.",

keywords = "Information Theoretic Clustering, Likelihood and Entropy estimator, Non-parametric",

author = "Hideitsu Hino and Noboru Murata",

year = "2014",

doi = "10.1007/978-3-319-11179-7_15",

language = "English",

isbn = "9783319111780",

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

publisher = "Springer Verlag",

pages = "113--120",

booktitle = "Artificial Neural Networks and Machine Learning, ICANN 2014 - 24th International Conference on Artificial Neural Networks, Proceedings",

note = "24th International Conference on Artificial Neural Networks, ICANN 2014 ; Conference date: 15-09-2014 Through 19-09-2014",

}

TY - GEN

T1 - A non-parametric maximum entropy clustering

AU - Hino, Hideitsu

AU - Murata, Noboru

PY - 2014

Y1 - 2014

N2 - Clustering is a fundamental tool for exploratory data analysis. Information theoretic clustering is based on the optimization of information theoretic quantities such as entropy and mutual information. Recently, since these quantities can be estimated in non-parametric manner, non-parametric information theoretic clustering gains much attention. Assuming the dataset is sampled from a certain cluster, and assigning different sampling weights depending on the clusters, the cluster conditional information theoretic quantities are estimated. In this paper, a simple clustering algorithm is proposed based on the principle of maximum entropy. The algorithm is experimentally shown to be comparable to or outperform conventional non-parametric clustering methods.

AB - Clustering is a fundamental tool for exploratory data analysis. Information theoretic clustering is based on the optimization of information theoretic quantities such as entropy and mutual information. Recently, since these quantities can be estimated in non-parametric manner, non-parametric information theoretic clustering gains much attention. Assuming the dataset is sampled from a certain cluster, and assigning different sampling weights depending on the clusters, the cluster conditional information theoretic quantities are estimated. In this paper, a simple clustering algorithm is proposed based on the principle of maximum entropy. The algorithm is experimentally shown to be comparable to or outperform conventional non-parametric clustering methods.

KW - Information Theoretic Clustering

KW - Likelihood and Entropy estimator

KW - Non-parametric

UR - http://www.scopus.com/inward/record.url?scp=84958540640&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84958540640&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-11179-7_15

DO - 10.1007/978-3-319-11179-7_15

M3 - Conference contribution

AN - SCOPUS:84958540640

SN - 9783319111780

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

SP - 113

EP - 120

BT - Artificial Neural Networks and Machine Learning, ICANN 2014 - 24th International Conference on Artificial Neural Networks, Proceedings

PB - Springer Verlag

T2 - 24th International Conference on Artificial Neural Networks, ICANN 2014

Y2 - 15 September 2014 through 19 September 2014

ER -

A non-parametric maximum entropy clustering

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this