Finding the optimal number of features based on mutual information

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Citation (Scopus)

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.

LanguageEnglish
Title of host publicationAdvances in Fuzzy Logic and Technology 2017 - Proceedings of
Subtitle of host publicationEUSFLAT-2017 – The 10th Conference of the European Society for Fuzzy Logic and Technology, IWIFSGN’2017 – The 16th International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets
PublisherSpringer
Pages477-486
Number of pages10
Volume641
ISBN (Print)9783319668291
DOIs
StatePublished - 2018
Event10th Conference of the European Society for Fuzzy Logic and Technology, (EUSFLAT 2017) and 16th International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, IWIFSGN 2017, 11-15 September 2017, Warsaw, Poland - Warsaw, Poland
Duration: 11 Sep 201715 Sep 2017
http://www.eusflat2017.ibspan.waw.pl/

Publication series

NameAdvances in Intelligent Systems and Computing
Volume641
ISSN (Print)2194-5357

Conference

Conference10th Conference of the European Society for Fuzzy Logic and Technology, (EUSFLAT 2017) and 16th International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, IWIFSGN 2017, 11-15 September 2017, Warsaw, Poland
Abbreviated titleEUSFLAT2017
CountryPoland
CityWarsaw
Period11/09/1715/09/17
Internet address

Fingerprint

Feature extraction
Transparency

Keywords

  • Feature selection
  • Fuzzy models
  • Mutual information
  • Number of features

Cite this

Chen, P., Wilbik, A., van Loon, S., Boer, A-K., & Kaymak, U. (2018). Finding the optimal number of features based on mutual information. In Advances in Fuzzy Logic and Technology 2017 - Proceedings of: EUSFLAT-2017 – The 10th Conference of the European Society for Fuzzy Logic and Technology, IWIFSGN’2017 – The 16th International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets (Vol. 641, pp. 477-486). (Advances in Intelligent Systems and Computing; Vol. 641). Springer. DOI: 10.1007/978-3-319-66830-7_43
Chen, P. ; Wilbik, A. ; van Loon, S. ; Boer, A.-K. ; Kaymak, U./ Finding the optimal number of features based on mutual information. Advances in Fuzzy Logic and Technology 2017 - Proceedings of: EUSFLAT-2017 – The 10th Conference of the European Society for Fuzzy Logic and Technology, IWIFSGN’2017 – The 16th International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets. Vol. 641 Springer, 2018. pp. 477-486 (Advances in Intelligent Systems and Computing).
@inproceedings{25c449d083ce49fdaf2ed51cf549fa1b,
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.",
keywords = "Feature selection, Fuzzy models, Mutual information, Number of features",
author = "P. Chen and A. Wilbik and {van Loon}, S. and A.-K. Boer and U. Kaymak",
year = "2018",
doi = "10.1007/978-3-319-66830-7_43",
language = "English",
isbn = "9783319668291",
volume = "641",
series = "Advances in Intelligent Systems and Computing",
publisher = "Springer",
pages = "477--486",
booktitle = "Advances in Fuzzy Logic and Technology 2017 - Proceedings of",
address = "Germany",

}

Chen, P, Wilbik, A, van Loon, S, Boer, A-K & Kaymak, U 2018, Finding the optimal number of features based on mutual information. in Advances in Fuzzy Logic and Technology 2017 - Proceedings of: EUSFLAT-2017 – The 10th Conference of the European Society for Fuzzy Logic and Technology, IWIFSGN’2017 – The 16th International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets. vol. 641, Advances in Intelligent Systems and Computing, vol. 641, Springer, pp. 477-486, 10th Conference of the European Society for Fuzzy Logic and Technology, (EUSFLAT 2017) and 16th International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets, IWIFSGN 2017, 11-15 September 2017, Warsaw, Poland, Warsaw, Poland, 11/09/17. DOI: 10.1007/978-3-319-66830-7_43

Finding the optimal number of features based on mutual information. / Chen, P.; Wilbik, A.; van Loon, S.; Boer, A.-K.; Kaymak, U.

Advances in Fuzzy Logic and Technology 2017 - Proceedings of: EUSFLAT-2017 – The 10th Conference of the European Society for Fuzzy Logic and Technology, IWIFSGN’2017 – The 16th International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets. Vol. 641 Springer, 2018. p. 477-486 (Advances in Intelligent Systems and Computing; Vol. 641).

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Finding the optimal number of features based on mutual information

AU - Chen,P.

AU - Wilbik,A.

AU - van Loon,S.

AU - Boer,A.-K.

AU - Kaymak,U.

PY - 2018

Y1 - 2018

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.

KW - Feature selection

KW - Fuzzy models

KW - Mutual information

KW - Number of features

UR - http://www.scopus.com/inward/record.url?scp=85029416210&partnerID=8YFLogxK

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

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

M3 - Conference contribution

SN - 9783319668291

VL - 641

T3 - Advances in Intelligent Systems and Computing

SP - 477

EP - 486

BT - Advances in Fuzzy Logic and Technology 2017 - Proceedings of

PB - Springer

ER -

Chen P, Wilbik A, van Loon S, Boer A-K, Kaymak U. Finding the optimal number of features based on mutual information. In Advances in Fuzzy Logic and Technology 2017 - Proceedings of: EUSFLAT-2017 – The 10th Conference of the European Society for Fuzzy Logic and Technology, IWIFSGN’2017 – The 16th International Workshop on Intuitionistic Fuzzy Sets and Generalized Nets. Vol. 641. Springer. 2018. p. 477-486. (Advances in Intelligent Systems and Computing). Available from, DOI: 10.1007/978-3-319-66830-7_43