Music Recommendation System on Spotify Using Deep Learning
{BE Students}* Pranali Balkrushna Chopade, Nikita Baban Chormale, Pratiksha Pratap Shinde, Vishakha Suryakant Shitole.
{Faculty}** Sachin Sukhadev Bere, Dinesh Bhagwan Hanchate
Department of Computer Engineering, Dattakala Group of Institutions Faculty of Engineering, Swami- Chincholi Bhigwan-413130, University of Pune, Maharashtra, INDIA.
{Email}* Chopadep2003@gmail.com, nikitachormale4242@gmail.com, Shindepratikshapratap9@gmail.com, vishakhashitole19@gmail.com
{Email}** dineshbhanchate@gmail.com , ssbere.foe@dattakala.edu.in, sakadam.foe@dattakala.edu.in
Abstract
Music recommendation systems play a critical role in personalizing the listening experience for users, particularly in platforms like Spotify. This paper presents a deep learning-based music recommendation system aimed at improving the accuracy and relevance of song suggestions. By leveraging user interaction data, such as listening history, song preferences, and behavioral patterns, the proposed model employs deep neural networks (DNN) to capture complex relationships between users and music content. Specifically, a hybrid approach combining collaborative filtering and content-based filtering is utilized, with deep learning models enhancing the representation of both user profiles and song features. The system integrates audio embeddings, metadata, and user behavior signals to generate more personalized recommendations. Experimental results demonstrate that the deep learning-based approach outperforms traditional recommendation methods, offering better prediction accuracy and user engagement. This system has the potential to enhance user satisfaction by providing more relevant and diverse music suggestions, thus improving overall user experience on Spotify. This paper provides a detailed review of music recommendation systems, particularly those employed by Spotify. With the proliferation of deep learning techniques, the quality of recommendations has drastically improved. We analyze various models used by Spotify and similar platforms, review deep learning approaches in the context of music recommendation, and discuss the challenges and future trends in the field.
