Koneoppimisen perusteet (3cr)

Objective

After completing the course, the student is familiar with the core concepts of machine learning, the most common learning paradigms, and the operating principles of selected fundamental algorithms. The student understands the phases of the machine learning process from data preprocessing to model evaluation and is able to apply machine learning methods to practical problems using the Python programming language and common data analysis libraries. In addition, the course introduces example implementations of algorithms without relying on ready-made solutions provided by libraries.

After completing the course, the student is able to:

- identify different types of machine learning problems and select an appropriate method for each
- preprocess data for use in machine learning models
- understand and implement the operation of key machine learning algorithms both theoretically and in practice
- utilize ready-made libraries (e.g. scikit-learn) for training and evaluating machine learning models
- assess model performance and interpret results critically

Content

The specific selection and emphasis of algorithms depends on the course implementation, but the course includes a mix of models based on different operating principles, such as probabilistic models (Naive Bayes), tree-based models (Decision Trees), distance-based models (k-NN, k-Means), as well as algorithms useful for data processing, such as models related to coordinate transformations or dimensionality reduction (PCA).