Discovery of distinctive patterns in music

Abstract

This paper proposes a new view of pattern discovery in music: inductive querying a corpus for maximally general distinctive patterns. A pattern is distinctive if it is over-represented with respect to an anticorpus, and maximally general distinctive if no subsuming pattern is also distinctive. An algorithm for maximally general distinctive pattern discovery is presented and applied to folk song melodies from three geographic regions, and to chord sequences from three music genres. Distinctive patterns are applicable to a wide range of music analysis tasks where an anticorpus can be defined and contrasted with an analysis corpus.