In data mining, subgroup discovery refers to finding interesting subsets of data or a population with an interesting property. In a medical context, a simple (but interpretable) disease prediction model might for example perform mediocre on the overall population, but achieve excellent accuracy on a patient subgroup such as female patients with an age below 50. Discovering such subgroups would enable the exploitation of simple models for subsets of the population. Subgroup discovery is also related to fairness since a disease prediction model is expected to perform equally well for each patient. We offer a range of bachelor’s and master theses in this context with application ranging from biomedicine to environmental modeling.