Chemometrics

Introduction to the "concept of chemometrics": In this section the students are given an overview of the chemometric concept and how chemometrics works as a guiding principle in (i) the definition of objectives, (ii) the planning of experiments, (iii) the creation of information-rich data, (iv) modelling and evaluation, (v) the visualisation  of large amounts of data, and in (vi) validation and prediction. The foundations of (chemical) data analysis: In this section the focus is on the model concept and variability, and how models of variability can be used in data anaylsis. Experiment design: This section addresses how experimental design can be used so that data contains information, how this data can be analysed and evaluated and how this philosophy can be used to optimise (chemical) systems and processes where many variables affect the outcome. Various types of experimental designs, and analyses and optimisation methods are addressed. Multivariate data analysis: This section addresses how large, complex quantities of data (tables) that are constructed of a large number of correlated variables can be analysed so that (i) an overview of multivariate data can be achieved, (ii) similarities and differences between tests can be detected and interpreted, (iii) the relationship between blocks (tables) of data can be modelled and interpreted. Specific applications are addressed, such as: quantitative structure property relationships, multivariate calibration, multivariate classification, as well as the monitoring and control of industrial and other processes. Various types of multivariate projection methods, such as principal component analysis, partial least squares projections to latent structures (PLS) and orthogonal projections to latent structures (OPLS) are presented.