DocumentCode :
2706805
Title :
Feature Selection on Supervised Classification Using Wilks Lambda Statistic
Author :
El Ouardighi, A. ; El Akadi, A. ; Aboutajdine, D.
Author_Institution :
Univ. Hassan I, Settat
fYear :
2007
fDate :
28-30 March 2007
Firstpage :
51
Lastpage :
55
Abstract :
Variable and feature selection have become the focus of much research in areas of application for which datasets with tens or hundreds of thousands of variables are available. This paper addresses the feature selection problem for supervised classification. We operate this feature selection step by step, which leads to search for a criterion to quantify the most relevant variable and its contribution compared to the others already selected. In this article we present a feature selection method based on the Wilk´s lambda criterion which is a statistical one used in discriminant analysis. Our objective is to evaluate the performances of this method when used in another application different from its classical one i.e. the discriminant analysis. This criterion is compared to other very known algorithms in the field of the feature selection on various real data sets. The obtained results with the criterion of Wilk´s lambda are satisfactory and even better in some cases.
Keywords :
data mining; feature extraction; pattern classification; statistical analysis; discriminant analysis; feature selection; lambda criterion; lambda statistic; supervised classification; variable selection; Computational complexity; Computational intelligence; Data visualization; Filters; Mutual information; Performance analysis; Performance evaluation; Power generation economics; Search methods; Statistics; Datamining; Feature selection; Supervised classification; Wilks Lambda; discriminant power;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computational Intelligence and Intelligent Informatics, 2007. ISCIII '07. International Symposium on
Conference_Location :
Agadir
Print_ISBN :
1-4244-1158-0
Electronic_ISBN :
1-4244-1158-0
Type :
conf
DOI :
10.1109/ISCIII.2007.367361
Filename :
4218394
Link To Document :
بازگشت