University of Twente Student Theses

Login
As of Friday, 8 August 2025, the current Student Theses repository is no longer available for thesis uploads. A new Student Theses repository will be available starting Friday, 15 August 2025.

Decentralizing Trust: Blockchain-Integrated Federated Learning for Trustworthy Financial Anomaly Detection

Westenbroek, Twan (2025) Decentralizing Trust: Blockchain-Integrated Federated Learning for Trustworthy Financial Anomaly Detection.

[img] PDF
1MB
Abstract:Financial institutions face barriers in collaborative anomaly detection due to data privacy and trust concerns. Federated learning (FL) enables joint model training without sharing raw data, but its reliance on a central aggregator introduces a single point of trust and potential failure. We propose a conceptual framework following a systematic literature review in which a permissioned blockchain replaces the central FL server, and smart contracts validate and coordinate model updates. This decentralized architecture enhances transparency and accountability, as all updates are immutably recorded in the blockchain ledger. We examine how smart contracts and blockchain consensus mechanisms coordinate FL updates, and analyze the trade-offs in system complexity and trust. Our findings suggest that integrating blockchain into FL can improve auditability and distribute trust, but incurs extra computational and communication overhead. This conceptual contribution lays the groundwork for future prototyping and evaluation.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Business & IT BSc (56066)
Link to this item:https://purl.utwente.nl/essays/107866
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page