Machine Learning for Fraud Detection in Banking Cyber security Performance Evaluation of Classifiers and Their Real-Time Scalability

Abstract

Context: With the rise of fraudulent activity in banking systems, effective fraud detection measures have become critical to safeguarding financial security and customer trust. Machine learning techniques are commonly used to improve detection capabilities, however their efficacy varies by performance parameter. Understanding these variances is crucial for optimizing fraud detection systems for real-world use. Objectives: Our main aim of this paper is to detect fraud in banking cybersecurity using machine learning classifiers. Materials/Methods: In this paper, we have used several machine learning classifiers to detect fraudulent behavior in the banking system. We used classifiers like: - Logistic Regression, Random Forest, Support Vector Machine, K-Nearest Neighbors, Gradient Boosting, AdaBoost, and Decision Tree. We have estimated these classifiers' performance metrics like: - accuracy, precision, recall, and F1-scoire. Result: From the experimental observation we found that RF gives the highest accuracy of 0.985, Recall at 0.985000 and the classifier KNN has the highest score at 0.988937. It indicates that RF is the best classifier in comparison to others for detecting fraud in the banking system Simultaneously the classifiers AdaBoost and Gradient Boosting provide good precision and AUC-ROC values

Topics & Concepts