Presenting a Bank Fraud Detection Model Based on Machine Learning Algorithms: An Empirical Comparison of Logistic Regression, Decision Tree, and Random Forest

Presenting a Bank Fraud Detection Model Based on Machine Learning Algorithms: An Empirical Comparison of Logistic Regression, Decision Tree, and Random Forest

Pourya Talebi Omran1

1) PhD student in Public Administration, Policy Making, Islamic Azad University, Sari Branch, Sari, Iran.

IOI Code : AAHC-AAKFDF
Publication : National Conference on Management, Commercialization, Knowledge-Based Economy and Investment for Production(impeb.ir)
Abstract :
    The transformation of the banking industry through digitization has undoubtedly transformed access and convenience for customers around the world. However, this paradigm shift has also brought new challenges, most notably in the area of ​​cybersecurity. Conventional rules-based strategies for fraud detection have failed to keep pace with the rapid evolution of cyber threats, spurring an increased interest in more adaptive approaches such as unsupervised learning. Furthermore, the COVID-19 pandemic has exacerbated the problem of banking fraud, as the widespread transition to online platforms and the increase in charitable fund-related transactions have created opportunities for cybercriminals to exploit. In response to these urgent concerns, this study examines the use of machine learning algorithms in the analysis and detection of fraudulent banking transactions. The scientific innovation of this research lies in the development of specific models for this purpose and the application of novel preprocessing techniques to increase the detection accuracy. This study provides promising results by utilizing a diverse set of algorithms including Random Forest, Nearest Neighbor (KNN), Naïve Bayes, Decision Trees, and Logistic Regression. In particular, the logistic regression and decision tree models show impressive values ​​of accuracy and area under the curve (AUC) of about 0.98, 0.97, and 0.95, 0.94, respectively. Given the pervasive nature of banking fraud in today’s digital society, the application of AI algorithms for fraud detection is a vital and timely step that enhances security and trust in the financial ecosystem.

Keywords : banking cybersecurity fraud detection machine learning digital transactions