A novel entropy-based approach to feature selection

Chia Hao Tu; Chunshien Li

doi:10.1007/978-3-319-54472-4_42

A novel entropy-based approach to feature selection

Chia Hao Tu, Chunshien Li^*

^*Corresponding author for this work

National Central University

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

Abstract

The amount of features in datasets has increased significantly in the age of big data. Processing such datasets requires an enormous amount of computing power, which exceeds the capability of traditional machines. Based on mutual information and selection gain, the novel feature selection approach is proposed. With Mackey-Glass, S&P 500, and TAIEX time series datasets, we investigated how good the proposed approach could perform feature selection for a compact subset of feature variables optimal or near optimal, through comparing the results by the proposed approach to those by the brute force method. With these results, we determine the proposed approach can establish a subset solution optimal or near optimal to the problem of feature selection with very fast calculation.

Original language	English
Title of host publication	Intelligent Information and Database Systems - 9th Asian Conference, ACIIDS 2017, Proceedings
Editors	Satoshi Tojo, Le Minh Nguyen, Ngoc Thanh Nguyen, Bogdan Trawinski
Publisher	Springer Verlag
Pages	445-454
Number of pages	10
ISBN (Print)	9783319544717
DOIs	https://doi.org/10.1007/978-3-319-54472-4_42
State	Published - 2017
Externally published	Yes
Event	9th Asian Conference on Intelligent Information and Database Systems, ACIIDS 2017 - Kanazawa, Japan Duration: 03 04 2017 → 05 04 2017

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10191 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	9th Asian Conference on Intelligent Information and Database Systems, ACIIDS 2017
Country/Territory	Japan
City	Kanazawa
Period	03/04/17 → 05/04/17

Bibliographical note

Publisher Copyright:
© Springer International Publishing AG 2017.

Keywords

Feature selection
Information entropy
Probability density estimation
Time series dataset

Access to Document

10.1007/978-3-319-54472-4_42

Cite this

Tu, C. H., & Li, C. (2017). A novel entropy-based approach to feature selection. In S. Tojo, L. M. Nguyen, N. T. Nguyen, & B. Trawinski (Eds.), Intelligent Information and Database Systems - 9th Asian Conference, ACIIDS 2017, Proceedings (pp. 445-454). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10191 LNAI). Springer Verlag. https://doi.org/10.1007/978-3-319-54472-4_42

Tu, Chia Hao ; Li, Chunshien. / A novel entropy-based approach to feature selection. Intelligent Information and Database Systems - 9th Asian Conference, ACIIDS 2017, Proceedings. editor / Satoshi Tojo ; Le Minh Nguyen ; Ngoc Thanh Nguyen ; Bogdan Trawinski. Springer Verlag, 2017. pp. 445-454 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{65b836223ca94394a60ca19b23016f0e,

title = "A novel entropy-based approach to feature selection",

abstract = "The amount of features in datasets has increased significantly in the age of big data. Processing such datasets requires an enormous amount of computing power, which exceeds the capability of traditional machines. Based on mutual information and selection gain, the novel feature selection approach is proposed. With Mackey-Glass, S&P 500, and TAIEX time series datasets, we investigated how good the proposed approach could perform feature selection for a compact subset of feature variables optimal or near optimal, through comparing the results by the proposed approach to those by the brute force method. With these results, we determine the proposed approach can establish a subset solution optimal or near optimal to the problem of feature selection with very fast calculation.",

keywords = "Feature selection, Information entropy, Probability density estimation, Time series dataset",

author = "Tu, {Chia Hao} and Chunshien Li",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2017.; 9th Asian Conference on Intelligent Information and Database Systems, ACIIDS 2017 ; Conference date: 03-04-2017 Through 05-04-2017",

year = "2017",

doi = "10.1007/978-3-319-54472-4_42",

language = "英语",

isbn = "9783319544717",

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

publisher = "Springer Verlag",

pages = "445--454",

editor = "Satoshi Tojo and Nguyen, {Le Minh} and Nguyen, {Ngoc Thanh} and Bogdan Trawinski",

booktitle = "Intelligent Information and Database Systems - 9th Asian Conference, ACIIDS 2017, Proceedings",

address = "德国",

}

Tu, CH & Li, C 2017, A novel entropy-based approach to feature selection. in S Tojo, LM Nguyen, NT Nguyen & B Trawinski (eds), Intelligent Information and Database Systems - 9th Asian Conference, ACIIDS 2017, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10191 LNAI, Springer Verlag, pp. 445-454, 9th Asian Conference on Intelligent Information and Database Systems, ACIIDS 2017, Kanazawa, Japan, 03/04/17. https://doi.org/10.1007/978-3-319-54472-4_42

A novel entropy-based approach to feature selection. / Tu, Chia Hao; Li, Chunshien.
Intelligent Information and Database Systems - 9th Asian Conference, ACIIDS 2017, Proceedings. ed. / Satoshi Tojo; Le Minh Nguyen; Ngoc Thanh Nguyen; Bogdan Trawinski. Springer Verlag, 2017. p. 445-454 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10191 LNAI).

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

TY - GEN

T1 - A novel entropy-based approach to feature selection

AU - Tu, Chia Hao

AU - Li, Chunshien

N1 - Publisher Copyright: © Springer International Publishing AG 2017.

PY - 2017

Y1 - 2017

N2 - The amount of features in datasets has increased significantly in the age of big data. Processing such datasets requires an enormous amount of computing power, which exceeds the capability of traditional machines. Based on mutual information and selection gain, the novel feature selection approach is proposed. With Mackey-Glass, S&P 500, and TAIEX time series datasets, we investigated how good the proposed approach could perform feature selection for a compact subset of feature variables optimal or near optimal, through comparing the results by the proposed approach to those by the brute force method. With these results, we determine the proposed approach can establish a subset solution optimal or near optimal to the problem of feature selection with very fast calculation.

AB - The amount of features in datasets has increased significantly in the age of big data. Processing such datasets requires an enormous amount of computing power, which exceeds the capability of traditional machines. Based on mutual information and selection gain, the novel feature selection approach is proposed. With Mackey-Glass, S&P 500, and TAIEX time series datasets, we investigated how good the proposed approach could perform feature selection for a compact subset of feature variables optimal or near optimal, through comparing the results by the proposed approach to those by the brute force method. With these results, we determine the proposed approach can establish a subset solution optimal or near optimal to the problem of feature selection with very fast calculation.

KW - Feature selection

KW - Information entropy

KW - Probability density estimation

KW - Time series dataset

UR - http://www.scopus.com/inward/record.url?scp=85018562768&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-54472-4_42

DO - 10.1007/978-3-319-54472-4_42

M3 - 会议稿件

AN - SCOPUS:85018562768

SN - 9783319544717

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

SP - 445

EP - 454

BT - Intelligent Information and Database Systems - 9th Asian Conference, ACIIDS 2017, Proceedings

A2 - Tojo, Satoshi

A2 - Nguyen, Le Minh

A2 - Nguyen, Ngoc Thanh

A2 - Trawinski, Bogdan

PB - Springer Verlag

T2 - 9th Asian Conference on Intelligent Information and Database Systems, ACIIDS 2017

Y2 - 3 April 2017 through 5 April 2017

ER -

Tu CH, Li C. A novel entropy-based approach to feature selection. In Tojo S, Nguyen LM, Nguyen NT, Trawinski B, editors, Intelligent Information and Database Systems - 9th Asian Conference, ACIIDS 2017, Proceedings. Springer Verlag. 2017. p. 445-454. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-54472-4_42

A novel entropy-based approach to feature selection

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this