An asymptotically optimal kernel combined classifier

被引:1
作者
Mojirsheibani, Majid [1 ]
Kong, Jiajie [1 ]
机构
[1] Calif State Univ Northridge, Dept Math, Northridge, CA 91330 USA
基金
美国国家科学基金会;
关键词
Kernel; Hamming distance; Smoothing parameter; Combined classifier; RANDOM FORESTS; REGRESSION; IMPROVE;
D O I
10.1016/j.spl.2016.07.017
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
A kernel ensemble classifier is developed for accurate classification based on several initial classifiers. A data-driven choice of the smoothing parameter of the kernel is considered and the resulting classifier is shown to be asymptotically optimal. Therefore, the proposed combined classifier asymptotically outperforms each individual classifier. (C) 2016 Elsevier B.V. All rights reserved.
引用
收藏
页码:91 / 100
页数:10
相关论文
共 23 条
[12]   Combining estimates in regression and classification [J].
LeBlanc, M ;
Tibshirani, R .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1996, 91 (436) :1641-1650
[13]   Random forests and adaptive nearest neighbors [J].
Lin, Yi ;
Jeon, Yongho .
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 2006, 101 (474) :578-590
[14]  
Mojirsheibani M, 1999, J AM STAT ASSOC, V94, P600
[15]   Aggregating classifiers via Rademacher-Walsh polynomials [J].
Mojirsheibani, Majid ;
Montazeri, Zahra .
JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION, 2015, 85 (06) :1187-1199
[16]   Ensemble-based classifiers [J].
Rokach, Lior .
ARTIFICIAL INTELLIGENCE REVIEW, 2010, 33 (1-2) :1-39
[18]   Super learner [J].
van der Laan, Mark J. ;
Polley, Eric C. ;
Hubbard, Alan E. .
STATISTICAL APPLICATIONS IN GENETICS AND MOLECULAR BIOLOGY, 2007, 6
[19]   STACKED GENERALIZATION [J].
WOLPERT, DH .
NEURAL NETWORKS, 1992, 5 (02) :241-259
[20]   METHODS OF COMBINING MULTIPLE CLASSIFIERS AND THEIR APPLICATIONS TO HANDWRITING RECOGNITION [J].
XU, L ;
KRZYZAK, A ;
SUEN, CY .
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS, 1992, 22 (03) :418-435