Sparsifying the Fisher linear discriminant by rotation

DOI10.1111/RSSB.12092zbMATH Open1414.62244arXiv1408.4475OpenAlexW2012029996WikidataQ36204852 ScholiaQ36204852MaRDI QIDQ5378138FDOQ5378138

Authors: Ning Hao, Bin Dong, Jianqing Fan 

Publication date: 12 June 2019

Published in: Journal of the Royal Statistical Society Series B: Statistical Methodology (Search for Journal in Brave)

Abstract: Many high dimensional classification techniques have been proposed in the literature based on sparse linear discriminant analysis (LDA). To efficiently use them, sparsity of linear classifiers is a prerequisite. However, this might not be readily available in many applications, and rotations of data are required to create the needed sparsity. In this paper, we propose a family of rotations to create the required sparsity. The basic idea is to use the principal components of the sample covariance matrix of the pooled samples and its variants to rotate the data first and to then apply an existing high dimensional classifier. This rotate-and-solve procedure can be combined with any existing classifiers, and is robust against the sparsity level of the true model. We show that these rotations do create the sparsity needed for high dimensional classifications and provide theoretical understanding why such a rotation works empirically. The effectiveness of the proposed method is demonstrated by a number of simulated and real data examples, and the improvements of our method over some popular high dimensional classification rules are clearly shown.

Full work available at URL: https://arxiv.org/abs/1408.4475

Recommendations

• Sparse linear discriminant analysis with applications to high dimensional low sample size data
• General sparse multi-class linear discriminant analysis
• Sparse linear discriminant analysis by thresholding for high dimensional data
• A procedure of linear discrimination analysis with detected sparsity structure for high-dimensional multi-class classification
• Sparse quadratic classification rules via linear dimension reduction

zbMATH Keywords

principal componentsclassificationlinear discriminant analysishigh dimensional dataequivariancerotate-and-solve procedure

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Cited In (11)

• Simultaneous variable selection and class fusion with penalized distance criterion based classifiers
• Dynamic linear discriminant analysis in high dimensional space
• Discriminant analysis on high dimensional Gaussian copula model
• Orthogonalized Fisher discriminant
• An Efficient Convex Formulation for Reduced-Rank Linear Discriminant Analysis in High Dimensions
• A \(U\)-classifier for high-dimensional data under non-normality
• Consistency of test-based method for selection of variables in high-dimensional two-group discriminant analysis
• Recipes for sparse LDA of horizontal data
• High-dimensional consistencies of KOO methods in multivariate regression model and discriminant analysis
• Optimal Whitening and Decorrelation
• Classification by likelihood accordance functions

This page was built for publication: Sparsifying the Fisher linear discriminant by rotation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5378138)