Recent trends in multi-block data analysis in chemometrics for multi-source data integration

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.