Kick-one-out-based variable selection method for Euclidean distance-based classifier in high-dimensional settings

Kick-one-out-based variable selection method for Euclidean distance-based classifier in high-dimensional settings
复制标题

高维环境下基于欧氏距离的分类器的踢一出变量选择方法

DOI:
10.1016/j.jmva.2021.104756
复制
发表时间:
2021
影响因子:
1.6
通讯作者:
Hyodo Masashi
Hyodo Masashi
中科院分区:
数学2区
文献类型:
--
作者:
Nakagawa Tomoyuki;Watanabe Hiroki;Hyodo Masashi

文献摘要

参考文献

被引文献

相似文献

提出了一种高维环境下基于欧氏距离的分类器变量选择方法。我们担心的是,当特征值中包含冗余变量时,基于欧氏距离的分类器的预期误分类概率(EPMC)可能会随着维度的增加而增加。首先,我们展示了仅具有非冗余变量的基于欧氏距离的分类器比具有所有变量的基于欧氏距离的分类器降低了渐近EPMC。接下来,我们得到了一个kick-one-out的变量选择方法,有助于减少EPMC和证明其一致性的变量选择在高维的背景下。最后,我们进行了Monte Carlo模拟研究,以检查所提出的选择方法的有限样本性能。我们的模拟结果表明,选择方法经常选择包含非冗余变量的集合。我们还观察到,从选定的变量构造的歧视规则减少EPMC比从所有变量构造的歧视规则。
This paper presents a variable selection method for the Euclidean distance-based classifier in high-dimensional settings. We are concerned that the expected probabilities of misclassification (EPMC) for the Euclidean distance-based classifier may be increasing with dimension when redundant variables are included in feature values. First, we show the Euclidean distance-based classifier with only non-redundant variables reduces asymptotic EPMC more than the Euclidean distance-based classifier with all variables. Next, we obtain a kick-one-out based variable selection method that helps reduce EPMC and prove its consistency in variable selection in the context of high dimensionality. Finally, we conduct a Monte Carlo simulation study to examine the finite sample performance of the proposed selection method. Our simulation results show that the selection method frequently selects the set containing non-redundant variables. We also observed that the discrimination rules constructed from the selected variables reduce EPMC more than the discrimination rules constructed from all variables.
DOI: 10.32917/hmj/1206129611
发表时间: 1988
影响因子: 0.2
作者:
R. Nishii;Z. Bai;P. R. Krishnaiah
通讯作者: P. R. Krishnaiah
线性判别规则的变量选择准则及其在高维大样本数据下的最优性
DOI: 10.1016/j.jmva.2013.10.005
发表时间: 2014
期刊: J. Multivar. Anal.
影响因子: --
作者:
Masashi Hyodo;T. Kubokawa
通讯作者: T. Kubokawa
DOI: 10.32917/hmj/1544238034
发表时间: 2018
影响因子: 0.2
作者:
Tomoyuki Nakagawa
通讯作者: Tomoyuki Nakagawa
DOI: 10.32917/hmj/1206133544
发表时间: 1983
影响因子: 0.2
作者:
Y. Fujikoshi
通讯作者: Y. Fujikoshi
根据错误率和赤池信息标准选择两组判别分析中的变量
DOI: 10.1016/0047-259x(85)90092-2
发表时间: 1985
影响因子: 1.6
作者:
Y. Fujikoshi
通讯作者: Y. Fujikoshi