Evaluasi Augmentasi Generatif dan Oversampling pada Berbagai Arsitektur Model untuk Deteksi Penipuan Kartu Kredit : Evaluation of Generative Augmentation and Oversampling Across Model Architectures for Credit Card Fraud Detection

Gilang Bagus Ramadhan; Riska Dhenabayu; Achmad Kautsar; Achmad Fitro; Nurul Fajrin Ariyani

doi:10.57152/malcom.v6i3.2821

Authors

Gilang Bagus Ramadhan Universitas Negeri Surabaya
Riska Dhenabayu Universitas Negeri Surabaya https://orcid.org/0000-0002-6530-2099
Achmad Kautsar Universitas Negeri Surabaya
Achmad Fitro Universitas Negeri Surabaya
Nurul Fajrin Ariyani Institut Teknologi Sepuluh Nopember https://orcid.org/0000-0001-7373-2235

DOI:

https://doi.org/10.57152/malcom.v6i3.2821

Keywords:

CTGAN, Deteksi Penipuan Kartu Kredit, Ketidakseimbangan Kelas, SMOTE, XGBoost, TabTransformer

Abstract

Deteksi penipuan kartu kredit pada data dengan ketidakseimbangan kelas ekstrem merupakan tantangan karena model klasifikasi cenderung bias terhadap kelas mayoritas sehingga menurunkan kemampuan mendeteksi transaksi penipuan. Metode oversampling tradisional seperti Synthetic Minority Over-sampling Technique (SMOTE) banyak digunakan, tetapi sering belum mampu merepresentasikan kompleksitas distribusi data tabular. Sebaliknya, pendekatan generatif seperti Conditional Tabular Generative Adversarial Network (CTGAN) berpotensi menghasilkan data sintetis yang lebih realistis, meskipun evaluasinya pada berbagai model klasifikasi masih terbatas. Penelitian ini bertujuan membandingkan efektivitas empat metode augmentasi data, yaitu SMOTE, Borderline-SMOTE, SMOTE-ENN, dan CTGAN, pada model Extreme Gradient Boosting (XGBoost), TabTransformer, dan Deep Autoencoder. Evaluasi dilakukan menggunakan metrik yang sensitif terhadap kelas minoritas, meliputi Receiver Operating Characteristic–Area Under the Curve (ROC-AUC), Area Under the Precision–Recall Curve (AUPRC), serta metrik pendukung lainnya. Hasil menunjukkan bahwa kombinasi XGBoost dan Borderline-SMOTE memberikan keseimbangan precision–recall terbaik dengan F1-score 0,9269 dan AUPRC 0,8306, sedangkan TabTransformer dengan CTGAN mencapai kemampuan diskriminasi tertinggi dengan ROC-AUC 0,9846 dan precision 0,9795. Temuan ini menunjukkan bahwa efektivitas augmentasi bergantung pada kesesuaian karakteristik data sintetis dengan mekanisme pembelajaran model serta memberikan implikasi praktis bagi pengembangan sistem deteksi penipuan yang andal pada data tidak seimbang

Downloads

Download data is not yet available.

References

The Nilson Report, “Payment card fraud losses approach $34 billion.,” 2025, [Online]. Available: https://www.globenewswire.com/news-release/2025/01/06/3004931/0/en/%0APayment-Card-Fraud-Losses-Approach-34-Billion.html.

I. Y. Hafez, A. Y. Hafez, A. Saleh, A. A. Abd El-Mageed, and A. A. Abohany, “A systematic review of AI-enhanced techniques in credit card fraud detection,” J. Big Data, vol. 12, p. 6, 2025, doi: 10.1186/s40537-024-01048-8.

S. M. Darwish, S. EL-Naggar, and S. M. Elkaffas, “Securing financial transactions: exploring the role of lightweight blockchain-enabled deep learning for fraud detection in FinTech systems,” Cybersecurity, vol. 9, p. 8, 2026, doi: 10.1186/s42400-025-00436-8.

E. Ileberi, “Performance Analysis of Cost-Sensitive Oversampling Mechanisms for Credit Card Fraud Detection,” IEEE Access, vol. 13, no. December, pp. 202655–202664, 2025, doi: 10.1109/ACCESS.2025.3637132.

F. Movahedi, R. Padman, and J. F. Antaki, “Adult?: Mechanical Circulatory Support Limitations of receiver operating characteristic curve on imbalanced data?: Assist device mortality risk scores,” J. Thorac. Cardiovasc. Surg., vol. 165, no. 4, pp. 1433-1442.e2, 2021, doi: 10.1016/j.jtcvs.2021.07.041.

H. Yin, Y. Teng, X. Zhang, and L. Yu, “Adaptive online bagging with hybrid sampling for non-stationary imbalanced data streams: A case study on credit card fraud detection,” Inf. Process. Manag., vol. 63, p. 104841, 2026, doi: 10.1016/j.ipm.2026.104841.

M. Tayebi and S. El Kafhali, “A novel approach based on XGBoost classifier and Bayesian optimization for credit card fraud detection,” Cyber Secur. Appl., vol. 3, p. 100093, 2025, doi: 10.1016/j.csa.2025.100093.

S. Lai, J. Wu, W. Wei, and C. Ye, “MiKTr: Multiscale knowledge fusion transformer for consumer fraud detection,” Inf. Fusion, vol. 127, p. 103779, 2025, doi: 10.1016/j.inffus.2025.103779.

H. Fanai and H. Abbasimehr, “A novel combined approach based on deep Autoencoder and deep classifiers for credit card fraud detection,” Expert Syst. Appl., vol. 217, p. 119562, 2023, doi: 10.1016/j.eswa.2023.119562.

O. Habibi, M. Chemmakha, and M. Lazaar, “Imbalanced tabular data modelization using CTGAN and machine learning to improve IoT Botnet attacks detection,” Eng. Appl. Artif. Intell., vol. 118, p. 105669, 2023, doi: 10.1016/j.engappai.2022.105669.

M. Imani, M. Joudaki, A. Bagheri, and H. R. Arabnia, “Why ROC-AUC Is Misleading for Highly Imbalanced Data: In-Depth Evaluation of MCC, F2-Score, H-Measure, and AUC-Based Metrics Across Diverse Classifiers,” Technologies, 2026, doi: 10.3390/technologies14010054.

M. Zavvar et al., “A hybrid deep learning framework using synthetic oversampling, autoencoder, convolutional neural networks, and an attention mechanism for credit card fraud detection,” J. Big Data, vol. 13, p. 21, 2026, doi: 10.1186/s40537-025-01331-2.

A. Al-dulaimi and N. Nnamoko, “Efficient treatment of class imbalance for credit card fraud detection Efficient treatment of class imbalance for credit card fraud detection,” 2025, doi: 10.36227/techrxiv.175693752.21183675/v1.

J. Teng, “ScienceDirect Financial Data Reduction and Information Retention Strategy based on Principal Component Analysis ( PCA ) Algorithm,” Procedia Comput. Sci., vol. 262, pp. 218–226, 2025, doi: 10.1016/j.procs.2025.05.047.

M. Kurniawan, H. Putra, C. A. Mintaraga, and Hidayaturrahman, “When Two are Better Than One: Synthesizing Heavily Unbalanced Data,” Procedia Comput. Sci., vol. 269, pp. 485–501, 2025, doi: 10.1016/j.procs.2025.08.301.

S. Matharaarachchi, M. Domaratzki, and S. Muthukumarana, “Machine Learning with Applications Enhancing SMOTE for imbalanced data with abnormal minority instances,” Mach. Learn. with Appl., vol. 18, no. December 2023, 2024.

R. J. Kuo, K.-W. Zheng, F. E. Zulvia, and T. Kuo, “Impurity-based borderline SMOTE with affinity propagation for imbalanced data classification,” Comput. Ind. Eng., vol. 212, p. 111720, 2026, doi: https://doi.org/10.1016/j.cie.2025.111720.

I. Aruleba and Y. Sun, “Machine Learning with Applications Enhanced credit risk prediction using deep learning and SMOTE-ENN resampling,” Mach. Learn. with Appl., vol. 21, no. April, p. 100692, 2025, doi: 10.1016/j.mlwa.2025.100692.

J. Chen, Y. Liang, J. Liu, and M. Zhou, “Temporal Transformer with Conditional Tabular GAN for Credit Card Fraud Detection: A Sequential Deep Learning Approach,” Mathematics, 2026, doi: 10.1016/j.aei.2026.104514.

R. Lingeswari and S. Brindha, “Online payments fraud prediction using optimized genetic algorithm based feature extraction and modified loss with XG boost algorithm for classification,” Swarm Evol. Comput., vol. 95, no. April, p. 101934, 2025, doi: 10.1016/j.swevo.2025.101934.

M. Duan, R. Du, and S. Ma, “Temporal Hypergraph Attentive Transformer for regional financial risk prediction: Integrating risk contagion dynamics and temporal pattern mining,” Alexandria Eng. J., vol. 133, pp. 137–156, 2025, doi: 10.1016/j.aej.2025.11.005.

S. Alharbi, “Aware Credit Card Fraud Detection Using Autoencoders and Generative Ensemble Learning,” Expert Syst., pp. 1–25, 2026, doi: 10.1111/exsy.70256.

R. H. Egami, A. A. A. El-mageed, M. Gafar, and A. A. Abohany, “An enhanced secretary bird-inspired multi-classifier framework for high-performance Credit Card Fraud Detection in imbalanced data,” Alexandria Eng. J., vol. 146, no. May, pp. 184–217, 2026, doi: 10.1016/j.aej.2026.04.061.

E. Richardson, R. Trevizani, J. A. Greenbaum, and H. Carter, “Article The receiver operating characteristic curve accurately assesses imbalanced datasets The receiver operating characteristic curve accurately assesses imbalanced datasets,” Patterns, vol. 5, no. 6, p. 100994, 2024, doi: 10.1016/j.patter.2024.100994.

S. Alharbi, K. Alahmadi, and X. Wang, “Imbalance-Aware Credit Card Fraud Detection Using Multi-Autoencoders and Generative Ensemble Learning,” Expert Syst., 2026, doi: 10.1111/exsy.70256.