From Exploration to Exploitation: Understanding the Evolution of Music Careers through a Data-driven Approach

• Danilo B. Seufitelli Universidade Federal de Minas Gerais http://orcid.org/0000-0002-0155-7631
• Mirella M. Moro Universidade Federal de Minas Gerais https://orcid.org/0000-0002-0545-2001

DOI: https://doi.org/10.5753/sbbd.2023.232399

Resumo

We propose a data-driven methodology to analyze how music artists spread their interest (explore) and focus their attention (exploit) on distinct music topics (genres) while having peaks of success (hot streaks). Music topics are identified through community detection over a complex network modeling of artists, songs and their genres. Then, we analyze exploration and exploitation by measuring the entropy and quantifying it in three periods: before, during and after hot streaks. Results show artists explore topics before hitting their first hot streak and during their most successful period. After, they tend to reduce the range of topics by going deeper in a handful only. Such findings are relevant for identifying and nurturing creative talent in the music industry.

Palavras-chave: artists' careers, musical success, entropy

Referências

Amorim, A. et al. (2022). Modelagem de tópicos em textos curtos: uma avaliação experimental. In Anais do XXXVII SBBD, pages 254–266, Búzios, RJ. SBC.

Baldo, F., Grando, J., Weege, K., and Bonassa, G. (2022). Adaptive fast xgboost for binary classification. In Anais do XXXVII SBBD, pages 13–25, Búzios, RJ. SBC.

Blei, D. M., Ng, A. Y., and Jordan, M. I. (2003). Latent dirichlet allocation. Journal of Machine Learning Research, 3:993–1022.

Byrd, D. and Crawford, T. (2002). Problems of music information retrieval in the real world. Information Processing & Management, 38(2):249–272.

Cosimato, A. et al. (2019). The conundrum of success in music: Playing it or talking about it? IEEE Access, 7:123289–123298.

Črepinšek, M., Liu, S.-H., and Mernik, M. (2013). Exploration and exploitation in evolutionary algorithms: A survey. ACM Computing Surveys (CSUR), 45(3):1–33.

Keogh, E. J. and Pazzani, M. J. (2000). Scaling up dynamic time warping for datamining applications. In SIGKDD, pages 285–289. ACM.

Liu, L. et al. (2018). Hot streaks in artistic, cultural, and scientific careers. Nature, 559(7714):396–399.

Liu, L. et al. (2021). Understanding the onset of hot streaks across artistic, cultural, and scientific careers. Nature Communications, 12(1):5392.

Martín-Gutiérrez, D. et al. (2020). A multimodal end-to-end deep learning architecture for music popularity prediction. IEEE Access, 8:39361–39374.

Oliveira, G. P. et al. (2020). Detecting collaboration profiles in success-based music genre networks. In ISMIR, pages 726–732.

Pachet, F. (2011). Hit song science. In Tao Li, Mitsunori Ogihara, G. T., editor, Music Data Mining, chapter 10, pages 305–326. CRC Press, New York, NY, USA.

Passman, D. (2019). All You Need to Know About the Music Business: 10th Edition. Simon & Schuster, New York.

Seufitelli, D. B., Oliveira, G. P., Silva, M. O., Barbosa, G. R. G., Melo, B. C., Botelho, J. E., Melo-Gomes, L. d., and Moro, M. M. (2022). From compact discs to streaming: A comparison of eras within the Brazilian market. Revista Vórtex, 10(1):1–28.

Silva, M. O., Rocha, L. M., and Moro, M. M. (2019). Collaboration Profiles and Their Impact on Musical Success. In ACM SAC, pages 2070–2077, Limassol, Cyprus.

Yucesoy, B. and Barabási, A.-L. (2016). Untangling performance from success. EPJ Data Science, 5(1):17.