Citation:

Sushant Rajaram Thite, 2025. "Adaptive Online Fraud Detection: Comparative Study of Machine Learning, Deep Learning, and Hybrid Models with Concept Drift Simulation", ESP Journal of Engineering & Technology Advancements  5(4): 119-128.

Abstract:

The rapid expansion of online financial transactions has led to an increase in fraudulent activities, posing significant challenges for digital security systems. Detecting fraud in real-time is complicated by the evolving nature of fraud strategies, a phenomenon known as concept drift, where the statistical properties of transaction data change over time. Traditional static Machine Learning (ML) models often struggle to maintain accuracy in such dynamic environments.This research presents a comparative study of Machine Learning (ML), Deep Learning (DL), and Hybrid models for adaptive online fraud detection, focusing on their ability to detect fraud under concept drift. The objective is to evaluate models such as Decision Trees, Random Forests, XGBoost, Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Convolutional Neural Networks (CNNs), and hybrid architectures that combine multiple approaches. Performance metrics such as accuracy, F1-score, precision, recall, latency, and drift adaptability will be used for evaluation.The study uses publicly available datasets, including the Kaggle Credit Card Fraud Detection dataset, to simulate real-world transaction streams. Concept drift is artificially introduced to test model robustness. An interactive Streamlit dashboard will be developed to visualize real-time model performance and drift detection over time.By comparing different types of models in both static and dynamic data environments, this study aims to identify techniques that balance accuracy, computational efficiency, and adaptability. The findings are expected to offer valuable insights for financial institutions, fintech companies, and cybersecurity professionals in developing real-time fraud detection systems that are both accurate and responsive to changing fraud tactics.

References:

[1] M. Niu et al., "Supervised vs Unsupervised Credit Card Fraud Detection (conference / arXiv)," arXiv:1904.10604.

[2] K. Shah & N. Mehta, "Comparative Study of Machine Learning Based Classification Techniques for Credit Card Fraud Detection," 2021.

[3] A. E-Arefin, "Comparative Study of ML Classifiers for Credit Card Fraud Detection," 2020.

[4] M. Makineedi et al., "Comparative Analysis of ML Models for Credit Card Fraud Detection," 2024. (searchable) [5] S. Verma and J. Dhar, "Credit Card Fraud Detection: A Deep Learning Approach," arXiv preprint arXiv:2409.13406, 2024. Available: https://arxiv.org/abs/2409.13406

[5] J. J. Assabil and I. C. Obagbuwa, “Credit Card Fraud Detection Using Machine Learning Algorithms: A Comparative Study of Six Models,” Int. J. Intelligent Systems and Applications in Engineering, vol. 12, no. 23s, pp. 862–?, Aug. 2024. Available: https://www.ijisae.org/index.php/IJISAE/article/view/7040

[6] C. Meng, L. Zhou, and B. Liu, “A Case Study in Credit Fraud Detection With SMOTE and XGBoost,” in J. Phys.: Conf. Ser., vol. 1601, no. 5, 2020, Art. no. 052016, doi: 10.1088/1742-6596/1601/5/052016.

[7] J. Taylor et al., "Comparative Analysis of Fraudulent Banking Transaction Models," 2025. (searchable) [9] Y. Lucas and J. Jurgovsky, "Credit Card Fraud Detection Using Machine Learning: A Survey," arXiv preprint arXiv:2010.06479, 2020. Available: https://arxiv.org/abs/2010.06479

[8] R. Randhawa et al., "AdaBoost and Voting Ensembles for Fraud Detection," IEEE Access, 2018.

[9] O. Ileberi et al., "Performance Evaluation of ML Methods Using SMOTE and AdaBoost," IEEE Access, 2021. Available: https://ieeexplore.ieee.org/

[10] M. Nishitha B S, “Fraud Detection in Card Transactions via Random Forest,” Int. J. Multidisciplinary Res., vol. 7, no. 4, pp. –, Jul.–Aug. 2025, doi: 10.36948/ijfmr.2025.v07i04.53825.

[11] G. Aburbeian and H. Ashqar, "Credit Card Fraud Detection Using Enhanced Random Forest Classifier for Imbalanced Data," arXiv preprint arXiv:2303.06514, 2023. Available: https://arxiv.org/abs/2303.06514

[12] S. Verma and J. Dhar, "Credit Card Fraud Detection: A Deep Learning Approach," arXiv preprint arXiv:2409.13406, 2024. Available: https://arxiv.org/abs/2409.13406

[13] S. Sulaiman et al., "Credit Card Fraud Detection Using Improved Deep Learning Models," CMC, 2024.

[14] S. Zahid, H. M. U. Hafeez, M. J. Iqbal, A. Asif, S. Yaqoob, and F. Mehboob, “Credit Card Fraud Detection using Deep Learning and Machine Learning Algorithms,” J. Innov. Comput. Emerg. Technol., vol. 4, no. 1, pp. –, Mar. 2024, doi: 10.56536/jicet.v4i1.106.

[15] T. Benchaji et al., "Enhanced Credit Card Fraud Detection Based on Attention + LSTM," 2021. Available: https://arxiv.org/

[16] Y. Chang et al., "Credit Card Fraud Detection Using Advanced Transformer Model," arXiv preprint arXiv:2406.03733, 2024. Available: https://arxiv.org/abs/2406.03733

[17] Q. Sha et al., "Detecting Credit Card Fraud via Heterogeneous Graph Neural Networks with Graph Attention," arXiv preprint arXiv:2504.08183, 2025. Available: https://arxiv.org/abs/2504.08183

[18] Y. Mazzer, G. Bontempi, et al., "GraphGuard: Contrastive Self-Supervised Learning for Credit-Card Fraud Detection in Multi-Relational Dynamic Graphs," arXiv preprint arXiv:2407.12440, 2024. Available: https://arxiv.org/abs/2407.12440

[19] M. Habibpour et al., "Uncertainty-Aware Credit Card Fraud Detection Using Deep Learning," arXiv preprint arXiv:2107.13508, 2021. Available: https://arxiv.org/abs/2107.13508

[20] E. Btoush et al., "A Hybrid Dependable Ensemble Machine Learning Model for Credit Card Fraud Detection," MDPI Applied Sciences, 2025. Available: https://www.mdpi.com/

[21] S. Gostkowski et al., "Credit Card Fraud Detection Using Machine Learning Techniques," 2024.

[22] A. Stotsky, "Development of New Methods for Detection and Control of Credit Card Fraud Attacks," arXiv preprint arXiv:2503.20477, 2025. Available: https://arxiv.org/abs/2503.20477

[23] D. Lunghi et al., "FRAUD-RLA: A Reinforcement Learning Adversarial Attack Against Credit Card Fraud Detection," arXiv preprint arXiv:2502.02290, 2025. Available: https://arxiv.org/abs/2502.02290

[24] I. Popova and H. Gardi, "Credit Card Fraud Detection: Comparative Benchmarking," arXiv preprint arXiv:2509.15044, 2025. Available: https://arxiv.org/abs/2509.15044

[25] E. Oztemel and M. Isik, "A Systematic Review of Intelligent Systems and Analytic Applications in Credit Card Fraud Detection," MDPI Applied Sciences, 2025. Available: https://www.mdpi.com/2076-3417/15/3/135

Keywords:

Credit Card Fraud Detection, Machine Learning, Deep Learning, Hybrid Model, Imbalanced Data, Comparative Study.