DocumentCode :
495291
Title :
Laplacian MinMax Discriminant Projections
Author :
Zhao, Jianmin ; Zheng, Zhonglong
Author_Institution :
Dept. of Comput. Sci., Zhejiang Normal Univ., Jinhua, China
Volume :
6
fYear :
2009
fDate :
March 31 2009-April 2 2009
Firstpage :
43
Lastpage :
47
Abstract :
A new algorithm, Laplacian minmax discriminant projection (LMMDP), is proposed in this paper for supervised dimensionality reduction. LMMDP aims at learning a linear transformation which is an extension of linear discriminant analysis (LDA). Specifically, we define the within-class scatter and the between-class scatter using similarities which are based on pairwise distances in sample space. After the transformation, the considered pairwise within the same class are as close as possible, while those between classes are as far as possible. The structural information of classes is contained in the within-class and the between-class Laplacian matrices. Thus the discriminant projection subspace can be derived by controlling the structural evolution of Laplacian matrices. The performance on several data sets demonstrates the competence of the proposed algorithm.
Keywords :
learning (artificial intelligence); matrix algebra; minimax techniques; pattern classification; statistical analysis; LDA; LMMDP algorithm; Laplacian matrix; Laplacian minmax discriminant projection; linear discriminant analysis; linear transformation; maximize between-class scatter; minimize within-class scatter; pattern classification; statistical analysis; supervised dimensionality reduction; supervised learning; Computer science; Feature extraction; Laplace equations; Linear discriminant analysis; Minimax techniques; Pattern recognition; Principal component analysis; Scattering; Unsupervised learning; Vectors; Dimensionality Reduction; Laplacian Matrix; Linear Discriminant Analysis; Supervised Learning;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Science and Information Engineering, 2009 WRI World Congress on
Conference_Location :
Los Angeles, CA
Print_ISBN :
978-0-7695-3507-4
Type :
conf
DOI :
10.1109/CSIE.2009.164
Filename :
5170658
Link To Document :
بازگشت