Performance and validation of MCR-ALS with quadrilinear constraint in the analysis of noisy datasets

详细信息    查看全文

• 作者：Amrita Malik ; ^{ambqam@cid.csic.es" class="auth_mail} ; ^{amritamalik.b@gmail.com" class="auth_mail} ; Roma Tauler ^{Tauler@idaea.csic.es" class="auth_mail rtaqam@cid.csic.es" class="auth_mail}
• 关键词：Four-way datasets ; MCR-ALS ; MLPCA ; Noisy data ; Quadrilinear constraint
• 刊名：Chemometrics and Intelligent Laboratory Systems
• 出版年：15 July 2014
• 年：2014
• 卷：135
• 期：Complete
• 页码：223-234
• 全文大小：643 K

文摘

This study explores the effect of noise propagation on the resolution capability of multivariate curve resolution-alternating least squares with a recently developed quadrilinearity constraint (MCR-ALS_Q). To investigate the effect of application of the quadrilinearity constraint, four environmental profiles were simulated and three types of noise viz. homoscedastic, heteroscedastic, and constant-proportional noise at three different levels were added to the simulated dataset. The profiles recovered with MCR-ALS_Q were compared with the ones recovered by bilinear MCR-ALS (MCR-ALS_B). The effect of maximum likelihood principal component analysis (MLPCA) as a pre-processing step in MCR-ALS_Q (MLPCA–MCR-ALS_Q), and MCR-ALS_B (MLPCA–MCR-ALS_B) analysis was also studied and results were compared with ones obtained with MCR-ALS_Q and MCR-ALS_B models. The recovery and similarity of the resolved profiles with theoretical ones were assessed in terms of similarity coefficient (r²) and similarity angle (胃). The results of this study conclude that MCR-ALS_Q is appropriate to analyze four-way quadrilinear datasets, and that the use of MLPCA as a pre-processing step before MCR-ALS_Q improves the resolution profiles to a great extent even in the presence of high levels of noise (heteroscedastic, and constant-proportional).