TY - JOUR
T1 - Recent trends in multi-block data analysis in chemometrics for multi-source data integration
AU - Mishra, Puneet
AU - Roger, Jean Michel
AU - Jouan-Rimbaud-Bouveresse, Delphine
AU - Biancolillo, Alessandra
AU - Marini, Federico
AU - Nordon, Alison
AU - Rutledge, Douglas N.
N1 - Publisher Copyright:
© 2021 The Author(s)
Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.
PY - 2021/4
Y1 - 2021/4
N2 - In recent years, multi-modal measurements of process and product properties have become widely popular. Sometimes classical chemometric methods such as principal component analysis (PCA) and partial least squares regression (PLS) are not adequate to analyze this kind of data. In recent years, several multi-block methods have emerged for this purpose; however, their use is largely limited to chemometricians, and non-experts have little experience with such methods. In order to deal with this, the present review provides a brief overview of the multi-block data analysis concept, the various tasks that can be performed with it and the advantages and disadvantages of different techniques. Moreover, basic tasks ranging from multi-block data visualization to advanced innovative applications such as calibration transfer will be briefly highlighted. Finally, a summary of software resources available for multi-block data analysis is provided.
AB - In recent years, multi-modal measurements of process and product properties have become widely popular. Sometimes classical chemometric methods such as principal component analysis (PCA) and partial least squares regression (PLS) are not adequate to analyze this kind of data. In recent years, several multi-block methods have emerged for this purpose; however, their use is largely limited to chemometricians, and non-experts have little experience with such methods. In order to deal with this, the present review provides a brief overview of the multi-block data analysis concept, the various tasks that can be performed with it and the advantages and disadvantages of different techniques. Moreover, basic tasks ranging from multi-block data visualization to advanced innovative applications such as calibration transfer will be briefly highlighted. Finally, a summary of software resources available for multi-block data analysis is provided.
KW - Chemometrics
KW - Data fusion
KW - Incremental learning
KW - Orthogonalization
KW - Pre-processing fusion
UR - http://www.scopus.com/inward/record.url?scp=85100879122&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85100879122&partnerID=8YFLogxK
U2 - 10.1016/j.trac.2021.116206
DO - 10.1016/j.trac.2021.116206
M3 - Review article
AN - SCOPUS:85100879122
VL - 137
SP - 1
EP - 15
JO - Trends in Analytical Chemistry
JF - Trends in Analytical Chemistry
SN - 0165-9936
M1 - 116206
ER -