Enhancing Data Integrity in Federated Learning for Cloud Load Prediction Using Blockchain

Khan, Abdullah (2025) Enhancing Data Integrity in Federated Learning for Cloud Load Prediction Using Blockchain. Masters thesis, Dublin, National College of Ireland.

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

Abstract

Federated learning (FL) allows decentralized machine learning and maintains privacy of the data, but it can be attacked by adversarial techniques, like model poisoning. This study suggests incorporating the use of blockchain technology and FL to help increase the integrity of data and model strength in the forecasting of cloud loads. A verification layer ensures that accountability and transparency are maintained via the creation of a blockchain-based involvement of model updates in a tamper-evident ledger. Tests on the system were carried out with clean state, attacked state, and attacked-verified state. The clean federated model performed better on the basis of Gradient Boosting Regressor (MAE = 0.0898, RMSE = 0.1112, MAPE = 46.81%,R² = 0.9507). In the attacked model, the performance fell in a huge manner (MAE = 0.2353, RMSE = 0.2934, MAPE = 85.70%, R² = 0.6565), whereas the attacked-verified model showed a recovery (MAE = 0.0914, RMSE = 0.1173, MAPE = 53.30%, R² = 0.9451). Large datasets produced in the course of the training were stored and managed on AWS S3 with the help of which data synchronization became efficient, and their storage was also scaled. The results emphasize how the blockchain-verified FL framework can effectively preserve the accuracy and transparency of the model and reduce the adversarial risks, which were related to the creation of privacy-preserving and secure machine learning architectures.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Arun, Shreyas Setlur UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Computer software > Computer Security > Database security > Blockchains (Databases) T Technology > T Technology (General) > Information Technology > Computer software > Computer Security > Database security > Blockchains (Databases) Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4050 Electronic information resources > Databases > Distributed databases > Blockchains (Databases) T Technology > T Technology (General) > Information Technology > Cloud computing Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Cloud Computing
Depositing User:	Ciara O'Brien
Date Deposited:	31 Aug 2026 15:27
Last Modified:	31 Aug 2026 15:27
URI:	https://norma.ncirl.ie/id/eprint/9713

Actions (login required)

View Item