A PCA Based Unsupervised Feature Selection Algorithm

Author

Luo, Yihui ; Xiong, Shuchu ; Wang, Sichun

Author_Institution

Dept. of Inf., Hunan Univ. of Commerce, Changsha

fYear

2008

fDate

25-26 Sept. 2008

Firstpage

299

Lastpage

302

Abstract

Principal components analysis (PCA) is an important approach to unsupervised dimensionality reduction. However, principal components (PCs) are a set of new variables carrying no clear physical meanings and still require all the original variables. To deal with this problem, the PC dominant feature (PCDF) is defined. Then, feature selection using them is considered and a new algorithm for determining such PC dominant features is proposed. Experimental results show that using the principal components as the basis the new algorithm can find a good feature subset.

Keywords

data mining; data reduction; feature extraction; pattern classification; principal component analysis; unsupervised learning; PCA based unsupervised feature selection algorithm; data mining; machine learning; pattern classification; principal component analysis; principal component dominant feature; unsupervised dimensionality reduction; Business; Clustering algorithms; Data structures; Digital signal processing; Extraterrestrial measurements; Feature extraction; Genetics; Partitioning algorithms; Personal communication networks; Principal component analysis; PCA; feature reduction; unsupervised feature selection;

fLanguage

English

Publisher

ieee

Conference_Titel

Genetic and Evolutionary Computing, 2008. WGEC '08. Second International Conference on

Conference_Location

Hubei

Print_ISBN

978-0-7695-3334-6

Type

conf

DOI

10.1109/WGEC.2008.109

Filename

4637449