This research presents a music recommendation framework that combines Fuzzy C-means clustering and cosine similarity to address challenges in personalized music discovery. Traditional methods, relying on collaborative filtering, often face the "cold start" problem for new users and neglect song attributes. The proposed system utilises audio features from Spotify to enhance accuracy and diversity in recommendations. Fuzzy C-means clustering accommodates songs in multiple clusters simultaneously, offering a nuanced representation of user preferences. Cosine similarity refines recommendations within the selected cluster, ensuring alignment with the user's specified song characteristics. The model introduces the "Other Side of the Coin" feature, encouraging users to explore songs with opposing characteristics to what they usually listen to. The paper outlines the background study, model architecture, and applications across individual users, streaming platforms, and curators. Implemented using essential libraries and the Flask framework, the proposed system aims to provide a user-centric web application for a more engaging and personalized music discovery experience. This research contributes to the field by offering a comprehensive and innovative solution to challenges in existing music recommendation systems .
Singh et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: