Collaboration-Aware Hit Song Prediction

• Autor(a) principal: Silva, Mariana O.
• Data de Publicação: 2023
• Outros Autores: Oliveira, Gabriel P., Seufitelli, Danilo B., Moro, Mirella M.
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Journal on Interactive Systems
• Texto Completo: https://sol.sbc.org.br/journals/index.php/jis/article/view/3137
• topic: Hit Song Science; Hit Song Prediction; Music Information Retrieval; Music Data Mining; Machine Learning

In a streaming-oriented era, predicting which songs will be successful is a significant challenge for the music industry. Indeed, there are many efforts in determining the driving factors that contribute to a song’s success, and one potential solution could be incorporating artistic collaborations, as it allows for a wider audience reach. Therefore, we propose a multi-perspective approach that includes collaboration between artists as a factor for hit song prediction. Specifically, by combining online data from Billboard and Spotify, we tackle the problem as both classification and hit song placement tasks, applying five different model variants. Our results show that relying only on music-related features is not enough, whereas models that also consider collaboration features produce better results.