Optimizing the Sentiment Recognition in Spotify Playlists Through Ensemble-Based Approaches

Abstract

Spotify, with over 320 million monthly active users reported in 2020, offers a unique platform for data science and machine learning applications. This study leverages Spotify’s extensive music library of over 50 million songs to analyze the emotional tone of user-created playlists using machine learning algorithms. By employing advanced classification methods, including Random Forest, Decision Tree, and Support Vector Machines (SVM), the research compares their effectiveness in sentiment classification tasks. The Random Forest model achieved the highest test accuracy of 87%, closely followed by the Decision Tree model at 86%. These results highlight the potential of sentiment-informed data to enhance music recommendation systems by tailoring suggestions to users’ emotional preferences. This work not only contributes to the evolving domain of sentiment-aware recommendation models but also demonstrates the technical challenges and practical implications of applying machine learning in music streaming platforms. The study’s findings underscore the value of integrating emotional intelligence into recommendation algorithms to improve user engagement and satisfaction in digital music services.

Keywords

• Sentiment Analysis
• Spotify API
• Playlist
• Machine Learning
• Music Recommendation Systems

References

1. AbeL, F., Gao, Q., Houben, G. J. & Tao, K. (2011). Analyzing user modeling on Twitter for personalized news recommendations. User ModeLing, Adaption, and PersonaLization, 1-12. Springer. google scholar
2. BirdaL, R. G. (2024). The Influence of Air PoLLution Concentrations on SoLar Irradiance Forecasting Using CNN-LSTM-mRMR Feature Extraction. CMC-Computers Materials and Continua, 78(3), 4015-4028. google scholar
3. Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5-32. google scholar
4. Chen, T. & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD InternationaL Conference on KnowLedge Discovery and Data Mining, 785-794. google scholar
5. DeLbouys, R., Hennequin, R., PiccoLi, F., Royo-LeteLier, J. & MoussaLLam, M. (2018). Music mood detection based on audio and lyrics with deep neural net. Proceedings of the 19th InternationaL Society for Music Information RetrievaL Conference (ISMIR), 688-695. google scholar
6. Ferraro, S., Bogdanov, D., Yoon, S., Kim, Y., & Serra, X. (2018). Cross-cultural analysis of user behavior in music streaming services. Proceedings of the 19th InternationaL Society for Music Information RetrievaL Conference (ISMIR), 1-6. google scholar
7. Hu, M. & Liu, B. (2004). Mining and summarizing customer reviews. Proceedings of the 10th ACM SIGKDD InternationaL Conference on KnowLedge Discovery and Data Mining, 168-177. google scholar
8. Hutto, C. J., & GiLbert, E. (2014). VADER: A parsimonious rule-based model for sentiment analysis of social media text. Proceedings of the 8th InternationaL Conference on WebLogs and SociaL Media (ICWSM), 216-225. google scholar

9. Jiang, L., Yu, M., Zhou, M., Liu, X., & Zhao, T. (2011). Target-dependent Twitter sentiment classification. Proceedings of the 49th Annual Meeting of the Association for ComputationaL Linguistics: Human Language Technologies, 151-160. google scholar
10. Li, Y., Wang, X., Chen, J., & Xu, Z. (2023). Sentiment anaLysis on music recommendation: A convoLutionaL neuraL network approach. IEEE Transactions on Multimedia, 25(3), 689-700. google scholar
11. Liu, H., & Singh, P. (2004). ConceptNet: A practicaL commonsense reasoning tooLkit. BT Technology Journal, 22(4), 211-226. google scholar
12. MumcuoğLu, K. Y. (2007). Biotherapy laboratory protocol department of parasitology. IsraeL: Hebrew University-Hadassah MedicaL SchooL JerusaLem. google scholar
13. Pang, B. & Lee, L. (2008). Opinion mining and sentiment anaLysis. Foundations and Trends in Information Retrieval, 2(1-2), 1-135. google scholar
14. Pappas, N. & Popescu-BeLis, A. (2016). Sentiment analysis of user comments for one-class collaborative filtering over text-based recom-mendations. Proceedings of the 39th InternationaL ACM SIGIR Conference on Research and DeveLopment in Information RetrievaL, 773-776. google scholar
15. Sancar, Y. (2024). Enhanced CLassification of Skin Lesions Using Fine-Tuned MobiLeNet and DenseNet121 ModeLs with EnsembLe Learning. Erzincan University Journal of Science and Technology, 17(3), 870-883. google scholar
16. SchedL, M. (2016). The LFM-1b dataset for music recommendation and user modeling. Proceedings of the 2016 ACM on InternationaL Conference on MuLtimedia RetrievaL, 103-110. google scholar
17. SchedL, M., Knees, P., McFee, B., & Bogdanov, D. (2021). Music recommendation: State of the art and chaLLenges. IEEE Signal Processing Magazine, 38(3), 29-44. google scholar
18. Turney, P. D. (2002). Thumbs up or thumbs down? Semantic orientation applied to unsupervised classification of reviews. Proceedings of the 40th AnnuaL Meeting of the Association for ComputationaL Linguistics, 417-424. google scholar
19. Zhang, J., Li, H., Zhao, R., & Huang, Y. (2022). Comparative study of LSTM and GRU for music sentiment anaLysis. IEEE Transactions on Neural Networks and Learning Systems, 33(5), 2345-2356. google scholar