Applied computational data analysis

• Relate parts of the course to the student’s own project
• Evaluate cross validation and concepts such as overfitting
• Evaluate and apply sparse regression and classification models
• Evaluate and apply logistic regression and support vector machines
• Evaluate and apply Classificaiton and regression trees (CART)
• Evaluate and apply random forests, boosting and ensemble methods
• Evaluate and ainterpret sparse latent methods such as sparse principal component analysis
• Evalute and interpret a range of unsupervised decomposition methods
• Evaluate clustering methods
• Compare and choose between the above methods

Methods: Cross-validation, elastic net, sparse principal components, sparse discriminant analysis and Gaussian mixture analysis, logistic regression, support vector machine, classification and regression trees, random forests, clustering, nonnegative matrix factorization, independent component analysis, sparse coding, archetypical analysis.

It is assumed that the participants have a basic knowledge of statistics or data analysis, and knowledge of Matlab, python or R.

One week of lectures and exercises. The lectures are given as flipped classroom with videos and Q&A sessions plus hands-on exercises. The participants work with a project on their own data and will hand in a report based on this after one month. Thus, it is expected that the participant work with their project after the teaching week.

Minimum: 8, Maximum: 60.

Please be aware that this course has a minimum requirement for the number of participants needed, in order for it to be held. If these requirements are not met, then the course will not be held. Furthermore, there is a limited number of seats available. If there are too many applicants, a pool will be created for the remainder of the qualified applicants, and they will be selected at random. You will be informed 8 days before the start of the course, whether you have been allocated a spot.