Self-Adaptive Federated Learning System for Financial Fraud Detection

Pananchickal Sebastian, Didheemose (2024) Self-Adaptive Federated Learning System for Financial Fraud Detection. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (945kB) | Preview

Abstract

Financial fraud is a major challenge for organisations looking forward to securing key information with strict adherence to privacy legislation. Traditional fraud detection approaches rely on a centralised mechanism that raises various privacy issues and always suffers from scalability and efficient resource utilization. Herein, this paper proposes a self-adaptive FL system for detecting financial fraud, allowing team-based training of models without actually sharing sensitive information. It adopts some intelligent methods, like model pruning and quantisation, for managing heterogeneous resources across clients, hence scalable. On the Kaggle Credit Card Fraud Detection dataset, the framework first prepares the data to fix the class imbalance and scale the features, hence strengthening the training. After the FL framework combines models with FedAvg, clustering methods are used to further improve model performance using clustering labels. Experimental results show increases in accuracy, precision, recall, and F1 score with reduced computational overhead in fraud detection. Current research points out opportunities of using FL in enhancing model performance on private datasets that are common in banks, health, and IoT devices using scalable and resource-efficient solutions.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Deshmukh, Sudarshan UNSPECIFIED
Uncontrolled Keywords:	Financial fraud; Self-adaptive Federated Learning (FL); Privacy issues; Scalability; Resource efficient solutions
Subjects:	H Social Sciences > HG Finance Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Cloud computing T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Telecommunications > Computer networks > Internet of things
Divisions:	School of Computing > Master of Science in Cloud Computing
Depositing User:	Ciara O'Brien
Date Deposited:	16 Jul 2025 09:49
Last Modified:	16 Jul 2025 09:49
URI:	https://norma.ncirl.ie/id/eprint/8133

Actions (login required)

View Item