Leveraging Machine Learning Models for Real-Time Fraud Detection in Financial Transactions

Nathaniel Andrew Davis; Sophia Anne Harris

doi:10.62951/ijcts.v1i1.56

Authors

Nathaniel Andrew Davis Massachusetts Institute of Technology (MIT)
Sophia Anne Harris Massachusetts Institute of Technology (MIT)

DOI:

https://doi.org/10.62951/ijcts.v1i1.56

Keywords:

Machine learning, fraud detection, financial transactions, real-time analysis, ensemble methods

Abstract

This study investigates the effectiveness of machine learning models in identifying fraudulent financial transactions in real-time. Using a large dataset of transactions, we compare the accuracy, precision, and speed of various models, including logistic regression, random forests, and neural networks. Our findings suggest that ensemble methods yield higher detection rates while minimizing false positives, thus providing a promising approach to financial fraud prevention.

References

Ahmed, E., Mahmood, A. N., & Hu, J. (2020). A survey of fraud detection techniques in financial transactions. Journal of Financial Crime, 27(4), 1035-1050.

Association of Certified Fraud Examiners (ACFE). (2020). Report to the Nations: Global Study on Occupational Fraud and Abuse.

Chen, L., Zhang, Y., & Zhao, Y. (2021). Deep learning for fraud detection in financial transactions: A review. Expert Systems with Applications, 165, 113-123.

Liu, Y., Wang, J., & Zhang, Y. (2022). Ensemble methods for fraud detection in financial transactions: A comparative study. Journal of Financial Technology, 3(1), 45-62.

McKinsey & Company. (2021). The State of AI in Financial Services.

Silva, D. A. G. F., Costa, A. M. A., & Ramos, J. (2021). Machine learning for fraud detection: An overview. Computers & Security, 105, 102-112.

Zhang, J., Wang, X., & Liu, H. (2022). A comprehensive review of ensemble learning techniques for fraud detection. IEEE Transactions on Knowledge and Data Engineering, 34(4), 123-136.