Music Recommendation Navigator

Recommender systems have seamlessly integrated into our daily routines, supporting us in discovering new music or curating entire playlists that resonate with our unique tastes. Recommender systems have become important tools for personalization, offering a tailored list of recommendations for each user by drawing insights from individual preferences, past interactions, and various other factors.

This bachelor thesis aims to develop a user-friendly web interface for exploring music recommendations. We particularly interested in developing methods that showcase the features of each song and that allow for the user to control the recommendations, leveraging a rich array of features. This includes not only high-level audio attributes such as danceability or tempo but also user-generated tags and emotion profiles associated with each song.