Finding the Optimal Number of Features Based on Mutual Information

Peipei Chen; Anna Wilbik; Saskia van Loon; Arjen-Kars Boer; Uzay Kaymak

doi:10.1007/978-3-319-66830-7_43

Finding the Optimal Number of Features Based on Mutual Information

Peipei Chen^*, Anna Wilbik, Saskia van Loon, Arjen-Kars Boer, Uzay Kaymak

^*Corresponding author for this work

Research output: Contribution to conference › Paper › Academic

Abstract

For high dimensional data analytics, feature selection is an indispensable preprocessing step to reduce dimensionality and keep the simplicity and interpretability of models. This is particularly important for fuzzy modeling since fuzzy models are widely recognized for their transparency and interpretability. Despite the substantial work on feature selection, there is little research on determining the optimal number of features for a task. In this paper, we propose a method to help find the optimal number of feature effectively based on mutual information.

Original language	English
Pages	477-486
DOIs	https://doi.org/10.1007/978-3-319-66830-7_43
Publication status	Published - 2017
Externally published	Yes
Event	Proceedings of the Conference of the European Society for Fuzzy Logic and Technology, International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets - Warsaw, Poland Duration: 11 Sept 2017 → 15 Sept 2017

Conference

Conference	Proceedings of the Conference of the European Society for Fuzzy Logic and Technology, International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets
Abbreviated title	EUSFLAT 2017, IWIFSGN 2017
Country/Territory	Poland
City	Warsaw
Period	11/09/17 → 15/09/17

Access to Document

10.1007/978-3-319-66830-7_43

Cite this

Chen, P., Wilbik, A., Loon, S. V., Boer, A.-K., & Kaymak, U. (2017). Finding the Optimal Number of Features Based on Mutual Information. 477-486. Paper presented at Proceedings of the Conference of the European Society for Fuzzy Logic and Technology, International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, Warsaw, Poland. https://doi.org/10.1007/978-3-319-66830-7_43

@conference{20095fc313ea4fddbaeb7c1daa9c5775,

title = "Finding the Optimal Number of Features Based on Mutual Information",

abstract = "For high dimensional data analytics, feature selection is an indispensable preprocessing step to reduce dimensionality and keep the simplicity and interpretability of models. This is particularly important for fuzzy modeling since fuzzy models are widely recognized for their transparency and interpretability. Despite the substantial work on feature selection, there is little research on determining the optimal number of features for a task. In this paper, we propose a method to help find the optimal number of feature effectively based on mutual information.",

author = "Peipei Chen and Anna Wilbik and Loon, {Saskia van} and Arjen-Kars Boer and Uzay Kaymak",

note = "DBLP's bibliographic metadata records provided through http://dblp.org/search/publ/api are distributed under a Creative Commons CC0 1.0 Universal Public Domain Dedication. Although the bibliographic metadata records are provided consistent with CC0 1.0 Dedication, the content described by the metadata records is not. Content may be subject to copyright, rights of privacy, rights of publicity and other restrictions.; Proceedings of the Conference of the European Society for Fuzzy Logic and Technology, International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, EUSFLAT 2017, IWIFSGN 2017 ; Conference date: 11-09-2017 Through 15-09-2017",

year = "2017",

doi = "10.1007/978-3-319-66830-7_43",

language = "English",

pages = "477--486",

}

Chen, P, Wilbik, A, Loon, SV, Boer, A-K & Kaymak, U 2017, 'Finding the Optimal Number of Features Based on Mutual Information', Paper presented at Proceedings of the Conference of the European Society for Fuzzy Logic and Technology, International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, Warsaw, Poland, 11/09/17 - 15/09/17 pp. 477-486. https://doi.org/10.1007/978-3-319-66830-7_43

Finding the Optimal Number of Features Based on Mutual Information. / Chen, Peipei; Wilbik, Anna; Loon, Saskia van et al.
2017. 477-486 Paper presented at Proceedings of the Conference of the European Society for Fuzzy Logic and Technology, International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, Warsaw, Poland.

Research output: Contribution to conference › Paper › Academic

TY - CONF

T1 - Finding the Optimal Number of Features Based on Mutual Information

AU - Chen, Peipei

AU - Wilbik, Anna

AU - Loon, Saskia van

AU - Boer, Arjen-Kars

AU - Kaymak, Uzay

N1 - DBLP's bibliographic metadata records provided through http://dblp.org/search/publ/api are distributed under a Creative Commons CC0 1.0 Universal Public Domain Dedication. Although the bibliographic metadata records are provided consistent with CC0 1.0 Dedication, the content described by the metadata records is not. Content may be subject to copyright, rights of privacy, rights of publicity and other restrictions.

PY - 2017

Y1 - 2017

N2 - For high dimensional data analytics, feature selection is an indispensable preprocessing step to reduce dimensionality and keep the simplicity and interpretability of models. This is particularly important for fuzzy modeling since fuzzy models are widely recognized for their transparency and interpretability. Despite the substantial work on feature selection, there is little research on determining the optimal number of features for a task. In this paper, we propose a method to help find the optimal number of feature effectively based on mutual information.

AB - For high dimensional data analytics, feature selection is an indispensable preprocessing step to reduce dimensionality and keep the simplicity and interpretability of models. This is particularly important for fuzzy modeling since fuzzy models are widely recognized for their transparency and interpretability. Despite the substantial work on feature selection, there is little research on determining the optimal number of features for a task. In this paper, we propose a method to help find the optimal number of feature effectively based on mutual information.

U2 - 10.1007/978-3-319-66830-7_43

DO - 10.1007/978-3-319-66830-7_43

M3 - Paper

SP - 477

EP - 486

T2 - Proceedings of the Conference of the European Society for Fuzzy Logic and Technology, International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets

Y2 - 11 September 2017 through 15 September 2017

ER -