Classifying Spotify Song Trends with SoundSoar: Machine Learning Insights for Content Creation and Marketing

Exploring Musical Streaming Patterns Through Machine Learning

Authors

  • Victoria Grasso Full Sail University
  • Andreas Marpaung Full Sail University

DOI:

https://doi.org/10.32473/flairs.38.1.139012

Keywords:

Machine Learning, agent-based model, Feature Engineering, Classification

Abstract

SoundSoar applies machine learning to predict Spotify song popularity trends—“up,” “down,” or “stable”—using engineered audio features and historical data. Among eight tested classifiers, the MLP achieved 97.0% accuracy, while ensemble models like Random Forest consistently performed well (90.0%–91.0%). By leveraging a tailored dataset and diverse models, this approach improves upon broader trend prediction methods. Despite challenges from Spotify API changes in late 2024, our findings validate machine learning’s role in music trend forecasting and set the stage for future enhancements, such as custom sound analysis and cross-platform models.

Downloads

Published

14-05-2025

How to Cite

Grasso, V., & Marpaung, A. (2025). Classifying Spotify Song Trends with SoundSoar: Machine Learning Insights for Content Creation and Marketing : Exploring Musical Streaming Patterns Through Machine Learning. The International FLAIRS Conference Proceedings, 38(1). https://doi.org/10.32473/flairs.38.1.139012