文献类型：期刊文献

英文题名：Variable selection in partial least squares with the weighted variable contribution to the first singular value of the covariance matrix

作者：Lin, Weilu[1];Hang, Haifeng[1];Zhuang, Yingping[1];Zhang, Siliang[1]

机构：[1]East China Univ Sci & Technol, State Key Lab Bioreactor Engn, 130 Meilong Rd, Shanghai 200237, Peoples R China

年份：2018

卷号：183

起止页码：113

外文期刊名：CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS

收录：;EI(收录号：20235015206069);WOS:【SCI-EXPANDED(收录号:WOS:000453490700012)】；

基金：The authors thank the Initiative Fund for Young Researchers of ECUST (YF0157126) for the financial support. Authors also thank valuable suggestions from anonymous reviewers to improve the quality of the manuscript.

语种：英文

外文关键词：Informative variables; Interval variable selection; Partial least squares; Variable contribution; Maximal singular value; Spectroscopy

摘要：The selection of informative variables in partial least squares (PLS) is important in process analytical technology (PAT) applications in the pharmaceutical industry, for example, the calibration of spectrometers. In the past, numerous approaches have been proposed to select the variables in partial least squares. In this work, a new variable selection method for PLS with the weighted variable contribution (PLS-WVC) to the first singular value of the covariance matrix for each PLS component is proposed. Several variants of PLS-WVC with different weighting factors are proposed. One variant of PLS-WVC is equivalent to the PLS with variable importance in projection (PIS-VIP). However, the variants with the correlation between X(gamma)w(gamma), and Y(gamma)q(gamma) as the weighting factor are preferred based on the results of the simulation cases studies. The proposed PLS-WVCs are integrated with interval PLS (iPLS) further to select the informative wavelength intervals for spectroscopic modelling. The utility of the proposed WVC based variable selection methods in PIS is demonstrated with the real spectral data sets.