APPLICATION OF MACHINE LEARNING METHODS IN THE BANKING SECTOR TO INCREASE THE EFFICIENCY OF DECISION-MAKING

DOI: 10.31673/2412-4338.2022.027685

Abstract

This article examines the modern banking sector in the context of its transformation through the integration of machine learning (ML). This process has become crucial for identifying new avenues for the optimization and automation of financial transactions. In the initial research phase, a deep analysis of existing decision-making methodologies in banks was conducted, focusing on their role in current conditions and the potential to replace human involvement with algorithms. The evaluation and calculation of ML model efficiency revealed certain challenges, such as overfitting and a lack of transparency, but also highlighted significant advantages over traditional systems.
When selecting tools for the implementation and deployment of ML, a need for specialized tools tailored for the banking sector was identified, considering the specific nuances of this market. Defined efficiency-enhancing directions include developing new regularization methods, improving cross-validation techniques, and advancing explainable AI.
The study also emphasized the ethical concerns arising from ML implementation. Notably, primary attention was given to ensuring user data confidentiality and reducing algorithmic bias.
Based on the research results, a conclusion about the immense potential of machine learning in the banking sector was drawn. While certain challenges exist, the right approach can lead to the creation of more efficient, secure, and transparent systems, enhancing client trust.

Keywords: machine learning, banking sector, decision making, explainability, data confidentiality, bias, audit, ethical practices, loan approval.

References
1. A.K. Choudhury, R.K. Swain "Machine learning in banking risk management: A literature review," Risk Management, vol. 20, pp. 1-30, 2018.
2. M. Zeng, T. Le, A. Liu, G. Zhang, M. Hussain, H. Chen, Q. Tian, and Z. Liu, "A Comparative Study on Decision Tree Algorithms for Bank Loan Default Prediction," in Procedia Computer Science, vol. 147, pp. 400-407, 2019.
3. H. Baesens , R. Van Gestel , S. Viaene , M. Stepanova, and J. Suykens , "Benchmarking state-of-the-art classification algorithms for credit scoring," Journal of the Operational Research Society, vol. 54, pp. 627-635, 2003.
4. P.A. Estévez , M. Tesmer, C.A. Perez, J.M. Zurada , "Normalized Mutual Information Feature Selection," IEEE Transactions on Neural Networks, vol. 20, no. 2, pp. 189-201, 2009.
5. R.K. Swain "Machine Learning for Risk Management in Banking Sectors: A Comparative Analysis of Supervised Algorithms," International Journal of Computational Intelligence Studies, vol. 7, no. 1/2, p. 37, 2018.
6. Zhebka V., Gertsiuk M., Sokolov V., Malinov V., Sablina M. Optimization of Machine Learning Method to Improve the Management Efficiency of Heterogeneous Telecommunication Network / CEUR Workshop Proceedings, 2022, с. 149-155.
7. Malinov, V., Zhebka, V., Zolotukhina, O., Franchuk, T., Chubaievskyi, V. Biomining as an Effective Mechanism for Utilizing the Bioenergy Potential of Processing Enterprises in the Agricultural Sector / CEUR Workshop Proceedings, 2023, с. 223-230.
8. Anakhov P., Zhebka V., Bondarchuk A., Storchak K., Sablina M. Increasing the Reliability of a Heterogeneous Network using Redundant Means and Determining the Statistical Channel Availability Factor / CEUR Workshop Proceedings, 2023, p. 231–236