Journal
JOURNAL OF MULTIVARIATE ANALYSIS
Volume 111, Issue -, Pages 350-367Publisher
ELSEVIER INC
DOI: 10.1016/j.jmva.2012.04.011
Keywords
Canonical Correlations; Consistency; High Dimension Low Sample Size; Misclassification; Naive Bayes rule
Categories
Funding
- KAKENHI [23500350]
- Grants-in-Aid for Scientific Research [23244011, 23500350] Funding Source: KAKEN
Ask authors/readers for more resources
This paper is concerned with pattern recognition for 2-class problems in a High Dimension Low Sample Size (HDLSS) setting. The proposed method is based on canonical correlations between the predictors X and responses Y. The paper proposes a modified version of the canonical correlation matrix Sigma(-1/2)(X) Sigma(XY) Sigma(-1/2)(Y) which is suitable for discrimination with class labels Y in a HDLSS context. The modified canonical correlation matrix yields ranking vectors for variable selection, a discriminant direction and a rule which is essentially equivalent to the naive Bayes rule. The paper examines the asymptotic behavior of the ranking vectors and the discriminant direction and gives precise conditions for HDLSS consistency in terms of the growth rates of the dimension and sample size. The feature selection induced by the discriminant direction as ranking vector is shown to work efficiently in simulations and in applications to real HDLSS data. (C) 2012 Elsevier Inc. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available