About Filter Criteria for Feature Selection in Regression

Alexandra Degeest, Michel Verleysen, Benoît Frénay

Research output: Contribution in Book/Catalog/Report/Conference proceedingConference contribution

Abstract

Selecting the best group of features from high-dimensional datasets is an important challenge in machine learning. Indeed problems with hundreds of features have now become usual. In the context of filter methods, the selected relevance criterion used for filtering is the key factor of a feature selection method. To select an appropriate criterion among the numerous existing ones, this paper proposes a list of six necessary properties. This paper describes then three relevance criteria, the mutual information, the noise variance and the adjusted R-squared, and compares them in the view of the aforementioned properties. Any new, or popular, criterion could be analysed in the light of these properties.

Original languageEnglish
Title of host publicationAdvances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings
EditorsIgnacio Rojas, Gonzalo Joya, Andreu Catala
PublisherSpringer Verlag
Pages579-590
Number of pages12
ISBN (Print)9783030205171
DOIs
Publication statusPublished - 1 Jan 2019
Event15th International Work-Conference on Artificial Neural Networks, IWANN 2019 - Gran Canaria, Spain
Duration: 12 Jun 201914 Jun 2019

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11507 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference15th International Work-Conference on Artificial Neural Networks, IWANN 2019
CountrySpain
CityGran Canaria
Period12/06/1914/06/19

Fingerprint

Feature Selection
Learning systems
Feature extraction
Regression
Filter
Filter Method
Mutual Information
Machine Learning
High-dimensional
Filtering
Necessary
Relevance

Keywords

  • Feature selection
  • Regression
  • Relevance criteria

Cite this

Degeest, A., Verleysen, M., & Frénay, B. (2019). About Filter Criteria for Feature Selection in Regression. In I. Rojas, G. Joya, & A. Catala (Eds.), Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings (pp. 579-590). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11507 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-030-20518-8_48
Degeest, Alexandra ; Verleysen, Michel ; Frénay, Benoît. / About Filter Criteria for Feature Selection in Regression. Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings. editor / Ignacio Rojas ; Gonzalo Joya ; Andreu Catala. Springer Verlag, 2019. pp. 579-590 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{6d20a1bd05ae4c0a82a154d2c7761d41,
title = "About Filter Criteria for Feature Selection in Regression",
abstract = "Selecting the best group of features from high-dimensional datasets is an important challenge in machine learning. Indeed problems with hundreds of features have now become usual. In the context of filter methods, the selected relevance criterion used for filtering is the key factor of a feature selection method. To select an appropriate criterion among the numerous existing ones, this paper proposes a list of six necessary properties. This paper describes then three relevance criteria, the mutual information, the noise variance and the adjusted R-squared, and compares them in the view of the aforementioned properties. Any new, or popular, criterion could be analysed in the light of these properties.",
keywords = "Feature selection, Regression, Relevance criteria",
author = "Alexandra Degeest and Michel Verleysen and Beno{\^i}t Fr{\'e}nay",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/978-3-030-20518-8_48",
language = "English",
isbn = "9783030205171",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "579--590",
editor = "Ignacio Rojas and Gonzalo Joya and Andreu Catala",
booktitle = "Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings",
address = "Germany",

}

Degeest, A, Verleysen, M & Frénay, B 2019, About Filter Criteria for Feature Selection in Regression. in I Rojas, G Joya & A Catala (eds), Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11507 LNCS, Springer Verlag, pp. 579-590, 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Gran Canaria, Spain, 12/06/19. https://doi.org/10.1007/978-3-030-20518-8_48

About Filter Criteria for Feature Selection in Regression. / Degeest, Alexandra; Verleysen, Michel; Frénay, Benoît.

Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings. ed. / Ignacio Rojas; Gonzalo Joya; Andreu Catala. Springer Verlag, 2019. p. 579-590 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11507 LNCS).

Research output: Contribution in Book/Catalog/Report/Conference proceedingConference contribution

TY - GEN

T1 - About Filter Criteria for Feature Selection in Regression

AU - Degeest, Alexandra

AU - Verleysen, Michel

AU - Frénay, Benoît

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Selecting the best group of features from high-dimensional datasets is an important challenge in machine learning. Indeed problems with hundreds of features have now become usual. In the context of filter methods, the selected relevance criterion used for filtering is the key factor of a feature selection method. To select an appropriate criterion among the numerous existing ones, this paper proposes a list of six necessary properties. This paper describes then three relevance criteria, the mutual information, the noise variance and the adjusted R-squared, and compares them in the view of the aforementioned properties. Any new, or popular, criterion could be analysed in the light of these properties.

AB - Selecting the best group of features from high-dimensional datasets is an important challenge in machine learning. Indeed problems with hundreds of features have now become usual. In the context of filter methods, the selected relevance criterion used for filtering is the key factor of a feature selection method. To select an appropriate criterion among the numerous existing ones, this paper proposes a list of six necessary properties. This paper describes then three relevance criteria, the mutual information, the noise variance and the adjusted R-squared, and compares them in the view of the aforementioned properties. Any new, or popular, criterion could be analysed in the light of these properties.

KW - Feature selection

KW - Regression

KW - Relevance criteria

UR - http://www.scopus.com/inward/record.url?scp=85067548571&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-20518-8_48

DO - 10.1007/978-3-030-20518-8_48

M3 - Conference contribution

AN - SCOPUS:85067548571

SN - 9783030205171

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

SP - 579

EP - 590

BT - Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings

A2 - Rojas, Ignacio

A2 - Joya, Gonzalo

A2 - Catala, Andreu

PB - Springer Verlag

ER -

Degeest A, Verleysen M, Frénay B. About Filter Criteria for Feature Selection in Regression. In Rojas I, Joya G, Catala A, editors, Advances in Computational Intelligence - 15th International Work-Conference on Artificial Neural Networks, IWANN 2019, Proceedings. Springer Verlag. 2019. p. 579-590. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-20518-8_48