Secure Financial Transactions through Machine Learning Based Anomaly Detection and End to End Data Encryption

Abstract

The rapid expansion of digital financial services has introduced unprecedented transaction speed and accessibility, but it has also significantly increased exposure to cyber threats, fraud schemes, and data breaches. Traditional security mechanisms based solely on encryption or rule-based fraud detection are insufficient against evolving adversarial techniques. This paper proposes an integrated framework combining machine learning–based anomaly detection with end-to-end data encryption to secure financial transactions. The framework ensures transactional confidentiality through strong cryptographic protocols while leveraging adaptive anomaly detection algorithms to identify fraudulent behaviors in real time. Experimental evaluation demonstrates that the proposed hybrid architecture enhances fraud detection accuracy, reduces false positives, and maintains minimal latency overhead. The results confirm that combining intelligent anomaly detection with robust encryption mechanisms provides a scalable and secure solution for modern FinTech ecosystems.