Descriptive Subgroup Mining of Folk Music

Descriptive analysis of music corpora is important to musicologists who are interested in identifying the properties that characterize specific genres of music. In this study we present such an analysis of a large corpus of folk tunes, all labeled by their origin. Subgroup Discovery (SD) is a rule learning technique located at the intersection of predictive and descriptive induction. One of the advantages of using this technique is the intuitive and interpretable result in the form of a collection of simple rules. Classification accuracy is not the goal of this study. Instead, we discuss some of the highest scoring rules with respect to their descriptive power.