Implementasi Metode Clustering dengan Algoritma K-Means pada Sistem Rekomendasi Lagu berbasis Similaritas Fitur Audio

Naufal Ammanullah Arrasyid Yusuf; Ilham Saifudin; Ari Eko Wardoyo

doi:10.69693/ijmst.v4i3.12785

Authors

Naufal Ammanullah Arrasyid Yusuf Universitas Muhammadiyah Jember
Ilham Saifudin Universitas Muhammadiyah Jember https://orcid.org/0000-0002-2063-4524
Ari Eko Wardoyo Universitas Muhammadiyah Jember

DOI:

https://doi.org/10.69693/ijmst.v4i3.12785

Keywords:

Sistem Rekomendasi Lagu, K-Means, Ekstraksi Audio, Intra-List Similarity, Jaccard Similarity

Abstract

Sistem rekomendasi lagu berbasis Collaborative Filtering menghadapi keterbatasan fundamental berupa cold-start problem ketika berhadapan dengan lagu atau pengguna baru yang belum memiliki riwayat interaksi. Penelitian ini mengimplementasikan pendekatan Content-Based Filtering (CBF) menggunakan algoritma K-Means Clustering berbasis ekstraksi fitur audio untuk mengatasi permasalahan tersebut. Dataset yang digunakan terdiri dari 3.200 file lagu dalam format mp3, wav, aac, dan m4a yang dikumpulkan dari penyimpanan lokal, mencakup enam genre populer, yaitu Pop, Rock, Jazz, Electronic/EDM, Hip-Hop/Rap, dan Classical. Fitur audio diekstraksi menggunakan library Librosa, menghasilkan representasi numerik berupa Mel-Frequency Cepstral Coefficients (MFCC), Spectral Centroid, Zero Crossing Rate (ZCR), RMS Energy, Spectral Rolloff, Chroma STFT, Tempo, dan Beat Count. Setelah normalisasi Z-Score untuk menyeragamkan skala fitur, algoritma K-Means Clustering diterapkan dan dievaluasi menggunakan Within-Cluster Sum of Squares (WCSS) dan Calinski-Harabasz (CH) Index, menghasilkan partisi optimal pada k=6 yang selaras dengan representasi enam genre dalam dataset, dengan nilai WCSS 95,616,62 dan CH Index 85,92. Sistem rekomendasi menghasilkan output Top-N menggunakan dua metrik jarak, yaitu Euclidean Distance dan Cosine Similarity, yang dievaluasi menggunakan Overlap Coefficient, Jaccard Similarity, dan Intra-List Similarity (ILS). Hasil evaluasi menunjukkan konvergensi mutlak antara kedua metode pada Top-3 dengan nilai Jaccard Similarity 1,0, sementara rata-rata skor ILS stabil pada rentang 0,5 yang mengindikasikan keseimbangan optimal antara relevansi dan diversitas rekomendasi. Sistem diimplementasikan dalam aplikasi web berbasis Flask dengan visualisasi Diagram Venn dan heatmap sebagai pendukung evaluasi interaktif.

References

Bevec, M., Tkalčič, M., & Pesek, M. (2024). Hybrid music recommendation with graph neural networks. User Modeling and User-Adapted Interaction, 34(5), 1891–1928. https://doi.org/10.1007/s11257-024-09410-4

Bianchi, E. (2025). Beyond Relevance: An Adaptive Exploration-Based Framework for Personalized Recommendations. Faculty of Engineering, Free University of Bozen-Bolzano, NOI Techpark - via Bruno Buozzi, 1, Bozen-Bolzano, 39100, Italy. http://arxiv.org/abs/2503.19525

Chen, P., Dong, L., & Liu, Y. (2024). Design of music recommendation system based on EDA and K-means cluster analysis. Applied and Computational Engineering, 50(1), 317–323. https://doi.org/10.54254/2755-2721/50/20241695

Gul, M., & Rehman, M. A. (2023). Big data: an optimized approach for cluster initialization. Journal of Big Data, 10(1). https://doi.org/10.1186/s40537-023-00798-1

Mukhopadhyay, S., Kumar, A., Parashar, D., & Singh, M. (2024). Enhanced Music Recommendation Systems: A Comparative Study of Content-Based Filtering and K-Means Clustering Approaches. Revue d’Intelligence Artificielle, 38(1), 365–376. https://doi.org/10.18280/ria.380138

Nurhalizah, R. S., Ardianto, R., & Purwono, P. (2024). Analisis Supervised dan Unsupervised Learning pada Machine Learning: Systematic Literature Review. Jurnal Ilmu Komputer Dan Informatika, 4(1), 61–72. https://doi.org/10.54082/jiki.168

Oh, Y., Yun, S., Hyun, D., Kim, S., & Park, C. (2023). MUSE: Music Recommender System with Shuffle Play Recommendation Enhancement. Proceedings of the 32nd ACM International Conference on Information and Knowledge Management, CIKM ’23, 1928–1938. https://doi.org/10.1145/3583780.3614976

Pake, G. V. R., Mawardi, V. C., & Sutrisno, T. (2023). PENERAPAN METODE CONTENT BASED FILTERING DALAM IMPLEMENTASI SISTEM REKOMENDASI MUSIK. Jurnal Serina Sains, Teknik Dan Kedokteran, 1(2), 455–462. https://doi.org/10.24912/jsstk.v1i2.31037

Prey, R. (2020). Locating Power in Platformization: Music Streaming Playlists and Curatorial Power. Social Media and Society, 6(2). https://doi.org/10.1177/2056305120933291

Prey, R., Esteve Del Valle, M., & Zwerwer, L. (2022). Platform pop: disentangling Spotify’s intermediary role in the music industry. Information Communication and Society, 25(1), 74–92. https://doi.org/10.1080/1369118X.2020.1761859

Retnoningsih, E., & Pramudita, R. (2020). Mengenal Machine Learning Dengan Teknik Supervised dan Unsupervised Learning Menggunakan Python. BINA INSANI ICT JOURNAL, 7(2), 156–165. https://www.python.org/

Schedl, M., Zamani, H., Chen, C. W., Deldjoo, Y., & Elahi, M. (2022). Current challenges and visions in music recommender systems research. International Journal of Multimedia Information Retrieval, 7(2), 95–116. https://doi.org/10.1007/s13735-018-0154-2

Senapati, A., Kujur, N., & Yelleti, V. (2026). Dynamic Graph with Similarity-Aware Attention Graph Neural Network for Recommender Systems. Department of Computer Science and Engineering, SRM University Andhra Pradesh, Mangalagiri, Andhra Pradesh 522502, India. http://arxiv.org/abs/2605.05238

Srivastava, A. (2024). Feature-Based User-Centric Music Recommendation. Journal of Student Research Volume 13 Issue 4. www.JSR.org/hs

Tan, P. N., Karpatne, A., & Steinbach. (2022). Introduction to Data Mining (2nd ed.). Pearson Education.

Valkenborg, D., Rousseau, A. J., Geubbelmans, M., & Burzykowski, T. (2023). Unsupervised learning. In American Journal of Orthodontics and Dentofacial Orthopedics (Vol. 163, Number 6, pp. 877–882). Elsevier Inc. https://doi.org/10.1016/j.ajodo.2023.04.001

Visher Laja Jaja, Y., Susanto, B., Ricky Sasongko, L., & Kunci, K. (2020). Penerapan Metode Item-Based Collaborative Filtering Untuk Sistem Rekomendasi Data MovieLens. D’Cartesian: Jurnal Matematika Dan Aplikasi, Vol. 9, No. 2 (September 2020): 78-83. https://ejournal.unsrat.ac.id/index.php/decartesian

Widiyaningtyas, T., Saifudin, I., Zaeni, I. A. E., Maulana, M. Z. N., & Caesarendra, W. (2025). Memory-based collaborative filtering based on matrix factorization and Gower’s set rank. Journal of King Saud University - Computer and Information Sciences, 37(9). https://doi.org/10.1007/s44443-025-00261-6

Xinyue Li. (2024). Analysis of machine learning-based music recommendation system using Spotify datasets. Applied and Computational Engineering, 77(1), 49–55. https://doi.org/10.54254/2755-2721/77/20240650

Yuniar, P., Sitoena, J. K., Matius, D. M., Obed, G. B., Tinggi, S., & Jaffray, F. (2022). Sejarah Musik sebagai Dasar Pengetahuan dalam Pembelajaran Teori Musik. In Jurnal Musik dan Pendidikan Musik | (Vol. 3, Number 2).