AI-powered fraud detection in online banking: Using machine learning to improve security
DOI:
https://doi.org/10.59828/ijsrmst.v4i7.345Keywords:
Machine learning, anomaly detection, cybersecurity, explainable, AI (XAI), and AI-powered fraud detectionAbstract
This study looks at how machine learning (ML) and artificial intelligence (AI) might improve fraud detection in the online banking industry. Fraudsters are growing more skilled as more financial transactions shift to digital platforms, making the implementation of advanced security measures necessary. Machine learning models that analyze large datasets, detect anomalies, and lower the risk of financial fraud are used to assist AI-driven fraud detection systems. The author of this literature study critically assesses current AI/ML-based fraud detection techniques in terms of their efficacy, difficulties they confront, and potential pathways for scaling up their use as a solution. The paper highlights important developments in deep learning models, supervised and unsupervised learning, and anomaly detection methodology. The results demonstrate AI's potential to improve fraud detection accuracy while addressing algorithmic bias, data privacy, and adversarial assault. The study concludes by offering suggestions for improving the fraud detection system with regard to real-time fraud monitoring, Explainable AI (XAI), and incorporating blockchain technology into the security of digital banking.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 International Journal of Scientific Research in Modern Science and Technology

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.