Melody Characterization by a Fuzzy Rule System

TitleMelody Characterization by a Fuzzy Rule System
Publication TypeConference Paper
Year of Publication2008
Conference NameInternational Workshop on Machine Learning and Music, International Conference on Machine Learning
AuthorsPonce León, P., Rizo D., & Ramirez R.
AbstractWe present preliminary work on automatic human-readable melody characterization. In order to obtain such characterization, we extract a set of statistical descriptors from a dataset of MIDI files, apply a rule induction algorithm to obtain a set of (crisp) classification rules for melody track identification, and transform the crisp rules into fuzzy rules. We present and discuss some results.