Data models and methods, module 1

The first module familiarizes students with the basic machine learning concepts using the R programming language within the context of biomedical data analysis. The course aims to provide an intuitive overview for different learning schemes such that students are able to relate them with specific data analysis types. After completing the course, students should be able to: 1. Explain the purposes and limitations of common machine learning methods. 2. Understand the methodological concepts such as model training/testing, standardization, cross-validation, overfitting, the curse of dimensionality and regularization. 3. Know how to preprocess data and check the model assumptions in R. 4. Implement a simple machine learning workflow in R starting from data preparation.

24 h of lectures and related reading, 57 hours of exercises and assignments.

It is recommended that this course is taken after the courses “Bio-data”, “Data mining and data-based models”, “Elementary Programming” and “Mathematics for biomedical data analysis” from the minor programme in bioinformatics if these courses are in the student’s curriculum.

All the necessary study material is provided in the course materials.

All related learning assignments and exercises need to be returned and will be assessed by the responsible teacher. The assessment criteria are based on the learning outcomes of the course.

Juho Kontio

The course utilizes grading scale “Pass/ Fail”.

It is recommended that this course is taken after the courses “Bio-data”, “Data mining and data-based models”, “Basic Programming” and “Mathematics for biomedical data analysis” from the minor programme in bioinformatics if these courses are in the student’s curriculum. It is recommended that this course is taken after the introductory course and together or before the courses “Data analysis in practice, module 1” from the continuous education programme in bioinformatics and data analytics.

Web-based self-study in Moodle.