Segmentasi Perilaku Pengguna Kartu Kredit untuk Prediksi Default dengan K-Means Clustering
DOI:
https://doi.org/10.70052/jka.v4i2.1375Keywords:
K-Means Clustering, Segmentasi Pengguna, Kartu Kredit, Credit Default, Data MiningAbstract
Peningkatan penggunaan kartu kredit di berbagai lapisan masyarakat diikuti oleh meningkatnya risiko credit default yang berpotensi menimbulkan kerugian finansial bagi institusi keuangan. Penelitian ini bertujuan untuk melakukan segmentasi perilaku pengguna kartu kredit serta menganalisis potensi default pada setiap segmen menggunakan metode K-Means Clustering. Data yang digunakan berasal dari Credit Card User Behavior and Default Dataset dari Kaggle, yang mencakup 10.000 nasabah dengan 24 atribut. Metodologi penelitian meliputi praproses data, seleksi fitur berbasis korelasi, penentuan jumlah cluster optimal menggunakan Elbow Method dan Silhouette Analysis, serta visualisasi dengan Principal Component Analysis (PCA). Hasil penelitian menunjukkan bahwa jumlah cluster optimal diperoleh pada K=2 dengan nilai Silhouette Score sebesar 0,250, lebih tinggi dibandingkan K=3 (0,195) dan K=4 (0,178). Segmentasi yang dihasilkan mampu mengidentifikasi perbedaan karakteristik perilaku nasabah yang berkaitan dengan tingkat risiko default. Kontribusi penelitian ini terletak pada integrasi metode unsupervised learning untuk segmentasi perilaku dengan analisis risiko pada tiap cluster, serta pengembangan aplikasi berbasis web menggunakan Streamlit untuk mendukung prediksi segmen nasabah secara interaktif.
The increasing use of credit cards across various segments of society has led to a higher risk of credit default, potentially causing financial losses for financial institutions. This study aims to segment credit card user behavior and analyze default risk within each segment using the K-Means Clustering method. The dataset used is the Credit Card User Behavior and Default Dataset from Kaggle, consisting of 10,000 customers with 24 attributes. The research methodology includes data preprocessing, correlation-based feature selection, determination of the optimal number of clusters using the Elbow Method and Silhouette Analysis, and visualization using Principal Component Analysis (PCA). The results indicate that the optimal number of clusters is K=2, achieving the highest Silhouette Score of 0.250, compared to K=3 (0.195) and K=4 (0.178). The resulting segmentation identifies distinct behavioral characteristics associated with different levels of default risk. The main contribution of this study lies in integrating unsupervised learning for behavioral segmentation with risk analysis in each cluster, as well as implementing the model in a web-based application using Streamlit to support interactive decision-making.
References
A. Abdulhafedh, “Incorporating K-means, Hierarchical Clustering and PCA in Customer Segmentation,” J. City Dev. Vol. 3, 2021, Pages 12-30, vol. 3, no. 1, pp. 12–30, 2021, doi: 10.12691/jcd-3-1-3.
H. Ramadhan, M. R. Abdan Kamaludin, M. A. Nasrullah, and D. Rolliawati, “Comparison of Hierarchical, K-Means and DBSCAN Clustering Methods for Credit Card Customer Segmentation Analysis Based on Expenditure Level,” J. Appl. Informatics Comput., vol. 7, no. 2, pp. 246–251, 2023, doi: 10.30871/jaic.v7i2.5790.
F. D. S. Alhamdani, A. A. Dianti, and Y. Azhar, “Segmentasi Pelanggan Berdasarkan Perilaku Penggunaan Kartu Kredit Menggunakan Metode K-Means Clustering,” JISKA (Jurnal Inform. Sunan Kalijaga), vol. 6, no. 2, pp. 70–77, 2021, doi: 10.14421/jiska.2021.6.2.70-77.
A. H. Zikri and R. D. Risanty, “Segmentasi Pemegang Kartu Kredit melalui Klasterisasi K-Means Menggunakan Bahasa Pemrograman Python,” J. Sist. Informasi, Teknol. Inf. dan Komput., vol. 15, no. 3, pp. 490–503, 2025, [Online]. Available: https://jurnal.umj.ac.id/index.php/just-it/index
I. Muslim, K. Karo, A. Yusmanto, R. Setiawan, U. Telkom, and U. Surya, “Segmentation of Credit Card Customers Based on Their Credit Card Usage Behavior using The K-Means Algorithm,” vol. 2, no. 2, pp. 55–64, 2021, [Online]. Available: file:///D:/%23TRIDHARMA/%23PENELITIAN/Genap 2025-2026/Publikasi Faiz/referensi/4_Segmentation_of_Credit_Card_Customers_Based_on_The.pdf
R. Bhandary and B. K. Ghosh, “Credit Card Default Prediction: An Empirical Analysis on Predictive Performance Using Statistical and Machine Learning Methods,” J. Risk Financ. Manag., vol. 18, no. 1, 2025, doi: 10.3390/jrfm18010023.
F. Wahab, I. Khan, and S. Sabada, “Credit card default prediction using ML and DL techniques,” Internet Things Cyber-Physical Syst., vol. 4, no. June, pp. 293–306, 2024, doi: 10.1016/j.iotcps.2024.09.001.
E. Costa e Silva, I. C. Lopes, A. Correia, and S. Faria, “A logistic regression model for consumer default risk,” J. Appl. Stat., vol. 47, no. 13–15, pp. 2879–2894, 2020, doi: 10.1080/02664763.2020.1759030.
W. Cai, F. Yang, B. Yao, C. Li, and G. Sun, An adaptive k-means clustering algorithm based on grid and domain centroid weights for digital twins in the context of digital transformation, vol. 12, no. 1. Springer International Publishing, 2025. doi: 10.1186/s40537-025-01180-z.
A. A. Wani, “Comprehensive analysis of clustering algorithms: exploring limitations and innovative solutions,” PeerJ Comput. Sci., vol. 10, pp. 1–45, 2024, doi: 10.7717/PEERJ-CS.2286.
D. Arthur and S. Vassilvitskii, “K-means++: The advantages of careful seeding,” Proc. Annu. ACM-SIAM Symp. Discret. Algorithms, vol. 07-09-Janu, pp. 1027–1035, 2007.
Purnima Bholowalia and Arvind Kumar, “EBK-Means: A Clustering Technique based on Elbow Method and K-Means in WSN,” Int. J. Comput. Appl., vol. 105, no. 9, pp. 975–8887, 2014.
P. J. Rousseeuw, “Silhouettes: A graphical aid to the interpretation and validation of cluster analysis,” J. Comput. Appl. Math., vol. 20, no. C, pp. 53–65, 1987, doi: 10.1016/0377-0427(87)90125-7.
I. T. Jollife and J. Cadima, “Principal component analysis: A review and recent developments,” Philos. Trans. R. Soc. A Math. Phys. Eng. Sci., vol. 374, no. 2065, 2016, doi: 10.1098/rsta.2015.0202.
F. S. Wewengkang and A. Wibowo, “Analysis Of Online Loan Regional Clustering in Indonesia in 2024 Based On Outstanding And Default Rate ( TWP90 ) Using K-Means Clustering,” vol. 8, no. 1, pp. 2237–2246, 2026.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Faiz Daffa Kurnia, Supriatin, Dwi Nurani, Alfie Nur Rahmi (Author)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
