A Simple Evaluation Model for Feature Subset Selection Algorithms

Fuente: Redalyc
Guardado en:
Detalles Bibliográficos
Autor principal: Huei Diana Lee
Formato: Artículo científico
Lenguaje:en
Publicado: Asociación Española para la Inteligencia Artificial 2006
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1876426688185761792
author Huei Diana Lee
author_facet Huei Diana Lee
contents A Simple Evaluation Model for Feature Subset Selection Algorithms Huei Diana Lee Wu Feng Chung Richardson Floriani Voltolini Ronaldo Cristiano Prati Maria Carolina Monard Ingeniería Machine Learning Feature Selection Multicriteria Evaluation The aim of Feature Subset Selection – FSS – algorithms is to select a subset of features from the originalset of features that describes a data set according to some importance criterion. To accomplish this task,FSS removes irrelevant and/or redundant features, as they may decrease data quality and reduce several ofthe desired properties of classifiers induced by supervised learning algorithms. As learning the best subsetof features is an NP-hard problem, FSS algorithms generally use heuristics to select subsets. Therefore, it isimportant to empirically evaluate the performance of these algorithms. However, this evaluation needs to bemulticriteria, i.e., it should take into account several properties. This work describes a simple model we haveproposed to evaluate FSS algorithms which considers two properties, namely the predictive performance ofthe classifier induced using the subset of features selected by different FSS algorithms, as well as the reductionin the number of features. Another multicriteria performance evaluation model based on rankings, whichmakes it possible to consider any number of properties is also presented. The models are illustrated by theirapplication to four well known FSS algorithms and two versions of a new FSS algorithm we have developed. 2006 artículo científico 1137-3601 https://www.redalyc.org/articulo.oa?id=92503203 en http://www.redalyc.org/revista.oa?id=925 Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial application/pdf Asociación Española para la Inteligencia Artificial Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial (España) Num.32 Vol.10
format Artículo científico
id redalyc_92503203
institution Redalyc
language en
publishDate 2006
publisher Asociación Española para la Inteligencia Artificial
spellingShingle A Simple Evaluation Model for Feature Subset Selection Algorithms
Huei Diana Lee
Ingeniería
Machine Learning
Feature Selection
Multicriteria Evaluation
A Simple Evaluation Model for Feature Subset Selection Algorithms Huei Diana Lee Wu Feng Chung Richardson Floriani Voltolini Ronaldo Cristiano Prati Maria Carolina Monard Ingeniería Machine Learning Feature Selection Multicriteria Evaluation The aim of Feature Subset Selection – FSS – algorithms is to select a subset of features from the originalset of features that describes a data set according to some importance criterion. To accomplish this task,FSS removes irrelevant and/or redundant features, as they may decrease data quality and reduce several ofthe desired properties of classifiers induced by supervised learning algorithms. As learning the best subsetof features is an NP-hard problem, FSS algorithms generally use heuristics to select subsets. Therefore, it isimportant to empirically evaluate the performance of these algorithms. However, this evaluation needs to bemulticriteria, i.e., it should take into account several properties. This work describes a simple model we haveproposed to evaluate FSS algorithms which considers two properties, namely the predictive performance ofthe classifier induced using the subset of features selected by different FSS algorithms, as well as the reductionin the number of features. Another multicriteria performance evaluation model based on rankings, whichmakes it possible to consider any number of properties is also presented. The models are illustrated by theirapplication to four well known FSS algorithms and two versions of a new FSS algorithm we have developed. 2006 artículo científico 1137-3601 https://www.redalyc.org/articulo.oa?id=92503203 en http://www.redalyc.org/revista.oa?id=925 Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial application/pdf Asociación Española para la Inteligencia Artificial Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial (España) Num.32 Vol.10
title A Simple Evaluation Model for Feature Subset Selection Algorithms
topic Ingeniería
Machine Learning
Feature Selection
Multicriteria Evaluation
url https://www.redalyc.org/articulo.oa?id=92503203