With the rapid development of streaming media, the forecast of the music popularity has practical value for platform optimization and recommendation as well as for creators to grasp market tendency. Existing studies generally train models with limited features and overlook the inherent class imbalance and feature scarcity of this task, which tends to produce undesirably low Precision or Recall. To solve the problems above, we proposed BEAT model. Considering the influences of artists and music genres on music popularity, we newly increased corresponding aggregated features. To tackle class imbalance, higher weights are assigned to minority class samples, and grid search is employed to determine optimal hyper parameter values. We conduct comparative experiments against eight machine learning models including SVM, Random Forest, and K-Nearest Neighbors. Experimental results demonstrate that our model achieves state of the art performance in terms of Accuracy, AUC and F1-score, reaching 94.99%, 97.39% and 79.15% respectively. It also obtains competitive Precision and Recall of 78.50% and 79.81%. These results verify that our model delivers favorable performance for music popularity prediction.

Jiangbei Wang, Yuxiang Fei, Yifu Zhao, Wenbo Wang, Wenyi Xiao. BEAT: A Feature-Enhanced and Balanced Prediction Model for Music Popularity. Academic Journal of Computing & Information Science (2026), Vol. 9, Issue 8: 36-42. https://doi.org/10.25236/AJCIS.2026.090805.

[1] Moreau, F., Viotto, J., & Wikström, P. (2026). Consumer satisfaction with new content discovery through algorithmic and human recommendations. Applied Economics Letters, 1–5. https://doi.org/10.1080/13504851.2026.2655917

[2] Sangeetha, A., Tejashree, M. , &  Roshini, B.. (2025). Spotify Dataset: Recommendation of Popularity Songs (Genre) Using Machine Learning Classification Techniques. International Conference on Hybrid Intelligent Systems. Springer, Cham.

[3] Kamal, J., Priya, P., Anala, M. R., & Smitha, G. R. (2021, September). A classification based approach to the prediction of song popularity. In 2021 international conference on innovative computing, intelligent communication and smart electrical systems (ICSES) (pp. 1-5). IEEE.

[4] Velmuruga, D. A. K.. (2023). Machine learning approaches for predicting song popularity: a case study in music analytics. INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT, 07(11), 1-11.

[5] Dong, A. , Qiu, R., & Ye, Z.. (2023). Regression analysis of song popularity based on ridge, k-nearest neighbors and multiple-layers neural networks. Highlights in Science, Engineering and Technology.

[6] Jain, V. , Gupta, R. , Chauhan, A. , &  Dubey, E. . (2025). Cracking the Code of Spotify Hits: The Impact of Data Preprocessing and Machine Learning Models. 2025 14th International Conference on System Modeling & Advancement in Research Trends (SMART). IEEE.

[7] Shil, T. , Afrin, S., Hossain, N. , Saha, P., Maitra, P., & Anwar, M. . (2025). Predicting Song Popularity by Analyzing Audio Features of Spotify Bengali Tracks across Diverse Genres. 2025 International Conference on Electrical, Computer and Communication Engineering (ECCE). IEEE. pp1-6.