Course content

This course is an introduction to experimental design and machine learning methods for modeling and data analysis with emphasis on applications to chemistry, biotechnology, process chemistry, material science, and physics. The goal of the course is to provide knowledge on methods that can be used to extract useful information from complex data sets or physical processes. The course gives knowledge about the planning of efficient experiments (how can we set up and execute experiments so that we can get as much information with as few experiments as possible), modeling (how to build and validate robust models), and methods that can be used to extract and discover useful information in complex data sets. Specifically, the following themes will be covered: design and planning of experiments, analysis of variance, regression, modeling with validation, techniques from machine learning for solving problems such as classification, clustering, analysis of multivariate data sets, and the use of neural networks for modeling.

Learning outcome

The course gives knowledge about:

1. Planning of experiments with analysis and interpretation of results from experimental designs.
2. Important steps for building and validating models with an emphasis on prediction.
3. Use of unsupervised analysis methods such as principal component analysis, explorative cluster methods for the analysis and interpretation of large data sets.
4. Robust regression methods (e.g. partial least squares) for analysis, interpretation, and prediction for large data sets.
5. Use of elementary machine learning techniques for classification.
6. Use of neural networks for modeling.

Learning methods and activities

The course will be centered around 4 seminars (physical or digital). Each seminar consists of two days and includes lectures, group-based exercises, and discussions

Specific conditions

Recommended previous knowledge

Basic mathematics (including some linear algebra), statistics, and chemistry. 

Course materials

The course material will be defined at the start of the course.