Aarhus University Seal / Aarhus Universitets segl

Predicting the Product Life Cycle of Songs on the Radio: How Record Labels Can Manage Product Portfolios and Prioritise Artists by Using Machine Learning Techniques

Research output: Contribution to conferencePaperResearchpeer-review

Abstract. In terms of determining the success of a musical artist's song, there is a positive correlation of radio play success and music sales success. Therefore, being able to forecast the future plays of a song on the radio can serve as powerful risk management and product portfolio management tools for record labels and other stakeholders of a song. This research strives to predict the remaining product life cycle of a song on the radio after it has been played for one or two months. The best results were achieved using a k-d tree to calculate the songs the most similar to the test songs and use a Random Forest model to forecast radio plays. An 82.78% and 83.44% accuracy was achieved for the two time periods, respectively. This explorative research leads to over 4500 test metrics to find the best combination of models and pre-processing techniques. Other algorithms tested were KNN, MLP, and CNN. The features only consist of daily radio plays and use no musical features.
Original languageEnglish
Publication year15 Jul 2021
Publication statusPublished - 15 Jul 2021

    Research areas

  • Hit Song Science, Product Life Cycle, Machine Learning, Radio

See relations at Aarhus University Citationformats

ID: 208292279