Chemometrics
The goal of the course is to teach students how to perform multivariate sensor calibration. Students will become familiar with the use of statistical concepts in chemometric applications. Most attention will be given to the ideas underlying the different methods and the application of these methods to realistic examples. Theoretical considerations and equations will be limited to what is needed to have sufficient insight to properly use the methods. Most examples will be related to spectroscopy and analytical chemistry, but the scope is broader. By using a combination of lectures, computer sessions and take home assignments the students will really learn how to apply the chemometric methods.
Praktische info:
Inschrijven?
- Voorwaarden: Knowledge of basic concepts of statistics and linear algebra is required. Some notions of analytical chemistry, sensor technology and multivariate statistics are a plus.
- Prijs: €125-€750
Georganiseerd door:
The following aspects of chemometrics will be handled in this course:
- Classical modelling concepts for quantitative calibration: Classical Least Squares (CLS), Inverse Least Squares (ILS), Multivariate Linear Regression (MLR), Principle Component Regression (PCR) and Partial Least Squares (PLS).
- Necessary steps for the creation and successful deployment of calibrations; Selection of calibration standards and assessment of the reliability of the models: (Test set validation vs. Cross-validation, model statistics). Special attention will be given to the methods for the selection of the number of principle components or latent variables in the projection methods.
- Methods for data pre-processing with special attention for the phenomena of light scattering and instrument drift and the methods to deal with these phenomena: derivatives, standard normal variate (SNV), multiplicative signal correction (MSC) and extended multiplicative signal correction (EMSC).
- Variable selection in a chemometric context and some commonly used methods for this.
- Qualitative analysis in a chemometric context: discrimination and classification
- New trends in chemometrics such as functional data analysis and augmented classical least squares (ACLS).
Course Materials
The course material will be made available :
- Slides from the lectures
- Papers discussed in the lectures
- Software manual
Additional material (suggested):
- A user-friendly guide to Multivariate Calibration and Classification by Naes, Isaksson, Fearn and Davies, NIR Publications 2004
- Multivariate Calibration by Martens and Naes, 1989
Dates
13, 20, 27 October and 10, 17, 24 November 2024
- Each time from 09.00 to 12.00
Lesgever/spreker
Wouter Saeys
Wouter Saeys is Full Professor at the KU Leuven Department of Biosystems in Belgium, where he leads the Biophotonics group with a focus on applications in the AgroFood chain. He received his Masters degree in Bioscience Engineering (2002) and a PhD in Bioscience Engineering (2006) from the KU Leuven. He was a postdoctoral researcher at the School for Chemical Engineering and Advanced Materials of the University of Newcastle upon Tyne (UK) and at the Norwegian Food Research Institute – Matforsk (Ås, Norway). In general Wouter’s research deals with light transport modeling and optical characterization of biological materials, multivariate data analysis and digital agriculture. He is author of 200+ research articles (ISI).
Gerelateerde opleidingen
Preparation for AI: From Raw Data to Reliable Models | How to make your data AI-ready
Opleiding - Brugge - PUC - KU Leuven Continue