This project studies how to predict songs that geriatric patients are likely to sing along to during empathy concert performances, replacing manual trial-and-error selection with a data-driven ranking approach.
Project journal
Music Recommendation System Being Used for Actual Empathy ConcertsHow the completed system is informing song selection for live performances.
Update on the Classical DSP SystemProgress on signal features, matching, and the explainable recommendation pipeline.
A New Side of Recommendation: Classical DSPExploring melody, harmony, rhythm, and emotion through audio signal processing.
Introduction to Music Recommendation SystemsThe project problem, intended audience, and early research direction.
Proposed Product and Early StepsThe first system architecture and approach to ranking candidate songs.
Recommendation System UpdateIteration notes, evaluation work, and lessons from the developing model.