User menu

Accès à distance ? S'identifier sur le proxy UCLouvain

Comparison Between Filter Criteria for Feature Selection in Regression

Primary tabs

download
  • Open access
  • PDF
  • 524.81 K
Degeest, Alexandra [UCL] Verleysen, Michel [UCL] Frénay, Benoît [UNamur]

High-dimensional data are ubiquitous in regression. To obtain a better understanding of the data or to ease the learning process, reducing the data to a subset of the most relevant features is important. Among the different methods of feature selection, filter methods are popular because they are independent from the model. This paper focuses on which properties make a good filter criterion, in order to be able to select one among the numerous existing ones. Six properties are discussed and three relevance criteria are compared with respect to the aforementioned properties.

Document type Communication à un colloque (Conference Paper) – Présentation orale avec comité de sélection
Access type Accès libre
Publication date 2019
Language Anglais
Conference "ICANN 2019", Munich (du 17/09/2019 au 19/09/2019)
Journal information "Artificial Neural Networks and Machine Learning - ICANN 2019 28th International Conference on Artificial Neural Networks" - (2019)
Peer reviewed yes
issn 0302-9743
e-issn 1611-3349
Publisher Springer International
Publication status Publié
Affiliation UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique
Keywords Feature Selection ; Filter Criteria ; Regression
Links
Bibliographic reference Degeest, Alexandra ; Verleysen, Michel ; Frénay, Benoît. Comparison Between Filter Criteria for Feature Selection in Regression.ICANN 2019 (Munich, du 17/09/2019 au 19/09/2019). In: Artificial Neural Networks and Machine Learning - ICANN 2019 28th International Conference on Artificial Neural Networks, (2019)
Permanent URL http://hdl.handle.net/2078.1/220382