Incremental Linear Discriminant Analysis: A Fast Algorithm and Comparisons

被引:55
作者
Chu, Delin [1 ]
Liao, Li-Zhi [2 ]
Ng, Michael Kwok-Po [3 ]
Wang, Xiaoyan [1 ]
机构
[1] Natl Univ Singapore, Dept Math, Singapore 119077, Singapore
[2] Hong Kong Baptist Univ, Dept Math, Hong Kong, Hong Kong, Peoples R China
[3] Hong Kong Baptist Univ, Ctr Math Imaging & Vis, Dept Math, Hong Kong, Hong Kong, Peoples R China
基金
芬兰科学院;
关键词
Classification accuracy; computational complexity; incremental linear discriminant analysis (ILDA); linear discriminant analysis (LDA); DIMENSION REDUCTION;
D O I
10.1109/TNNLS.2015.2391201
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
It has always been a challenging task to develop a fast and an efficient incremental linear discriminant analysis (ILDA) algorithm. For this purpose, we conduct a new study for linear discriminant analysis (LDA) in this paper and develop a new ILDA algorithm. We propose a new batch LDA algorithm called LDA/QR. LDA/QR is a simple and fast LDA algorithm, which is obtained by computing the economic QR factorization of the data matrix followed by solving a lower triangular linear system. The relationship between LDA/QR and uncorrelated LDA (ULDA) is also revealed. Based on LDA/QR, we develop a new incremental LDA algorithm called ILDA/QR. The main features of our ILDA/QR include that: 1) it can easily handle the update from one new sample or a chunk of new samples; 2) it has efficient computational complexity and space complexity; and 3) it is very fast and always achieves competitive classification accuracy compared with ULDA algorithm and existing ILDA algorithms. Numerical experiments based on some real-world data sets demonstrate that our ILDA/QR is very efficient and competitive with the state-of-the-art ILDA algorithms in terms of classification accuracy, computational complexity, and space complexity.
引用
收藏
页码:2716 / 2735
页数:20
相关论文
共 29 条
[1]   Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling [J].
Alizadeh, AA ;
Eisen, MB ;
Davis, RE ;
Ma, C ;
Lossos, IS ;
Rosenwald, A ;
Boldrick, JG ;
Sabet, H ;
Tran, T ;
Yu, X ;
Powell, JI ;
Yang, LM ;
Marti, GE ;
Moore, T ;
Hudson, J ;
Lu, LS ;
Lewis, DB ;
Tibshirani, R ;
Sherlock, G ;
Chan, WC ;
Greiner, TC ;
Weisenburger, DD ;
Armitage, JO ;
Warnke, R ;
Levy, R ;
Wilson, W ;
Grever, MR ;
Byrd, JC ;
Botstein, D ;
Brown, PO ;
Staudt, LM .
NATURE, 2000, 403 (6769) :503-511
[2]  
[Anonymous], 2001, Pattern Classification
[3]   Partitioning-based clustering for Web document categorization [J].
Boley, D ;
Gini, M ;
Gross, R ;
Han, EH ;
Hastings, K ;
Karypis, G ;
Kumar, V ;
Mobasher, B ;
Moore, J .
DECISION SUPPORT SYSTEMS, 1999, 27 (03) :329-341
[4]   Document categorization and query generation on the World Wide Web using WebACE [J].
Boley, D ;
Gini, M ;
Gross, R ;
Han, EH ;
Hastings, K ;
Karypis, G ;
Kumar, V ;
Mobasher, B ;
Moore, J .
ARTIFICIAL INTELLIGENCE REVIEW, 1999, 13 (5-6) :365-391
[5]  
Burk F., 1998, LEBESGUE MEASURE INT
[6]   CHARACTERIZATION OF ALL SOLUTIONS FOR UNDERSAMPLED UNCORRELATED LINEAR DISCRIMINANT ANALYSIS PROBLEMS [J].
Chu, Delin ;
Goh, Siong Thye ;
Hung, Y. S. .
SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 2011, 32 (03) :820-844
[7]   A new and fast implementation for null space based linear discriminant analysis [J].
Chu, Delin ;
Thye, Goh Siong .
PATTERN RECOGNITION, 2010, 43 (04) :1373-1379
[8]  
Dettling M, 2002, GENOME BIOL, V3
[9]  
Fukunaga K., 2013, Introduction to Statistical Pattern Recognition
[10]   From few to many: Illumination cone models for face recognition under variable lighting and pose [J].
Georghiades, AS ;
Belhumeur, PN ;
Kriegman, DJ .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2001, 23 (06) :643-660