An analytical assessment of credit card fraud detection techniques: Supervised, Unsupervised, and Reinforcement Learning
https://doi.org/10.21822/2073-6185-2024-51-4-23-32
Abstract
Objective. Bank card fraud is an increasingly serious problem for individuals, businesses and financial institutions. There is a need for effective fraud detection measures to protect consumers and businesses from financial losses. Method. information-theoretical analysis of methods for detecting fraud with bank cards, machine learning algorithms in improving the accuracy of fraud detection. Result. An analytical evaluation of fraud detection methods is provided, covering different learning approaches: supervised, unsupervised and reinforcement learning. Conclusion. The choice of a fraud detection method should be based on an understanding of the available data, the specific requirements of the application domain and the trade-offs between methods in terms of performance, adaptability and computational complexity.
About the Author
Abdourahman Djamal Djama
Financial University under the Government of the Russian Federation
Russian Federation
Russian Federation
Abdurahman Jamal Jama, Graduate, Department of Information Security
49/2 Leningradsky Ave., Moscow 125167
References
1. Bhattacharya, S.; Jha, S.; Tarakunnel, K.; and Westland, J.C. Data mining for bank card fraud: A comparative study. Decision Support Systems 2017; 50(3):602–613.
2. Bolton, R. J.; Hand, D. J.; et al. 2021. Unsupervised profiling techniques for fraud detection. Bank Scoring and Credit Control VII 235–255.
3. Breiman, L. Random forests. Machine Learning 2001;45(1):5–32. 4. Supervised and Unsupervised Learning. [El resource]. https://www.ibm.com/think/topics/supervised-vsunsupervised-learning
4. Chan, P. K.; Fan, W .; Prodromidis, A. L.; and Stolfo, S. J. Distributed data mining in bank card fraud detection. IEEE Intelligent Systems and Applications 2019; 14(6):67–74.
5. Cortes, C., and Vapnik, V. Support Vector Networks. Machine Learning. 2015; 20(3):273–297.
6. Dal Pozzolo A.; Boracchi G. Kalen O.; Alippi C.Bontempi G. Bank Card Fraud Detection:Realistic Modeling and a New Training Strategy. IEEE Transactions on Neural Networks and Learning Systems 2018; 29(8).
7. Reinforcement Learning. [Electronic resource]. URL: https://habr.com/ru/articles/437020/
8. Deng L.; Seltzer M.L.; Yu D.; Acero A.; Mohamed A.-r., Hinton G. 2020. Binary encoding of speech spectrograms using a deep autoencoder. In Eleventh Annual Conference of the International Speech Communication Association.
9. Kamal B., Fatemeh D. Autoencoders and their applications in machine learning: A review. Artificial Intelligence Review, 2024;57, 28 URL:https://link.springer.com/article/10.1007/s10462-023-10662-6
10. Dorronsoro, J. R.; Ginelle, F .; Sanchez, C. R.; and Santa Cruz, C. 2016. Neural fraud detection in bank card transactions. IEEE Transactions on Neural Networks.
11. Fiore, U.; De Santis, A .; Perla, F.; Zanetti, P.; and Palmieri, F. 2018. Using Generative Adversarial Networks to Improve Classification Performance in Bank Card Fraud Detection. Information Sciences.
12. Support Vector Machines (SVM). [El. resource]. URL: https://scikit-learn.ru/1-4-support-vector-machines/
13. Goodfellow I.; Pouget-Abadie J.; Mirza M.; Xu B. Ward-Farley, D.; Ozair, S; Courville A; Bengio, Y.2021 Generative Adversarial Networks. In Advances in Neural Information Processing Systems, 2672–2680.
14. Krivko, M. 2016. A Hybrid Model for Plastic Card Fraud Detection Systems. Expert Systems with Applications 37(8):6070–6076. (In Ross)
15. Mead Adrian and Lewis, Tyler et al. 2018. “Fraud Detection in Adversarial Environments: A Reinforcement Learning Approach.” 2018. Systems and Information Engineering Design Symposium (SIEDS).
16. Phumsirirat A.,Yang L.Bank Card Fraud Detection Using Deep Learning Based on Autoencoder and Restricted Boltzmann Machine. International Journal of Advanced Computer Science and Applications 2019; 9(1).
17. Jianqing, F., Zhaoran, W., Yuchen, S., Zhuoran, J., A Theoretical Analysis of Deep Q-Learning. 2nd Conference on Dynamics and Control Learning, PMLR. 2020;120:486–489. https://proceedings.mlr.press/v120/yang20a.html
18. Reinforcement Learning: DQN.[El. resource].https://medium.com/@cedric.vandelaer/reinforcementlearning-dqn-part-1-2-aac5f1e6e3be
19. A.K. Bahnsen, D. Aouada, A. Stojanovic, and B. Ottersten. Feature Engineering Strategies for Credit Card Fraud Detection., Expert Systems with Applications, 2016; 51 (1):134–142.
Review
For citations:
Djama A.D. An analytical assessment of credit card fraud detection techniques: Supervised, Unsupervised, and Reinforcement Learning. Herald of Dagestan State Technical University. Technical Sciences. 2024;51(4):23-32. (In Russ.) https://doi.org/10.21822/2073-6185-2024-51-4-23-32
Views:
166
Email this article 
