Promoting Unfamiliar Music Through Data Science: MARS, the Music Affect Recommender System for Digital Library Engagement

Abstract This article describes a personal music recommender system designed to overcome the “cold start” problem that affects the visibility of recent works of composition and improvisation outside popular music. The approach builds upon the authors’ experimental results showing increased levels of engagement with unfamiliar music through a focus on acoustic properties and emotional response. Set in the context of a digital music library, the developed recommender feature provides two visual acoustic representations of a musical work and allows users to record their continuous affect responses. These interface elements are designed to invoke the behaviors observed in the authors’ experiments: i.e., to keep visitors engaged with the online resource for longer periods and motivate them to experience a broader range of music.