Securing digital payments and transactions using federated learning

View at publisher Summarise with VIRAT
Authors: Vikram Singh, Yashasvi Makin, Srikanth Yerra, Ravi Prakash Chaturvedi, Annu Mishra, Rajneesh Kumar Singh
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The widespread adoption of digital payment systems has posed a critical challenge surrounding the security and privacy of financial transactions. Traditional centralized machine learning (ML) models offered an effective tool for fraud detection and anomaly detection, but they also opened up privacy risks by utilizing raw transactional data and aggregating it on a centralized server. This article describes a Federated Learning (FL)-based approach to secure digital payments as it provides the ability to train a model across multiple client machines while not exposing sensitive user data during any part of the transaction. Additionally, we provide the ability to execute secure FL to collaboratively train a transaction anomaly detection model using distributed datasets from payment endpoints (e.g., mobile apps and point-of-sale (POS) systems). Federated secure learning is shown to include secure model aggregation and differential privacy properties that protect user data from data leaks and adversarial attacks. The results from the various experiments demonstrate that the accuracy of the federated model can be on par or better than the centralized version while providing the significant benefit of greater data confidentiality and reduced risk of regulatory compliance issues. We also show that federated secure learning is feasible within a financial context, thus the research opens the door to privacy-preserving frameworks around digital payments and scalable procedural architectures.

Fields of Study & Subjects

Privacy-Preserving Technologies in Data Blockchain Technology Applications and Security Cryptography and Data Security

Publication details

Journal / source
CRC Press eBooks
Pages
148–169
Publication date
25 March 2026
Language
EN
Document type
Book chapter