Identifying Emotion Segments in Music by Discovering Motifs in Physiological Data

Music can induce different emotions in people. We propose a system that can identify music segments which induce specific emotions from the listener. The work involves building a knowledge base with mappings between affective states (happiness, sadness, etc.) and music features (rhythm, chord progression, etc.). Building this knowledge base requires background knowledge from music and emotions psychology. Psychophysiological responses of a user, particularly, the blood volume pulse, are taken while he listens to music. These signals are analyzed and mapped to various musical features of the songs he listened to. A motif discovery algorithm used in data mining is adapted to analyze signals of physiological data. Motif discovery finds patterns in the data that indicate points of interest in the music. The different motifs are stored in a library of patterns and used to identify other songs that have similar musical content. Results show that motifs selected have similar chord progressions. Some of which include frequently used chords in western pop music.