Enhancing Federated Learning Security: Combating Clients’ Data Poisoning with Classifier Ensembles DOI
Arunava Roy,

Dipankar Dasgupta

Published: Oct. 28, 2024

Language: Английский

DI4IoT: A comprehensive framework for IoT device-type identification through network flow analysis DOI
Saurav Kumar, Manoj Kumar Das,

Sukumar Nandi

et al.

Internet of Things, Journal Year: 2025, Volume and Issue: unknown, P. 101599 - 101599

Published: April 1, 2025

Language: Английский

Citations

0
THE-TAFL: Transforming Healthcare Edge with Transformer-based Adaptive Federated Learning and Learning Rate Optimization DOI
Farhan Ullah, Nazeeruddin Mohammad, Leonardo Mostarda

et al.

Internet of Things, Journal Year: 2025, Volume and Issue: unknown, P. 101605 - 101605

Published: April 1, 2025

Language: Английский

Citations

0
Navigating the fusion of federated learning and big data: a systematic review for the AI landscape DOI

R Haripriya,

Nilay Khare, Manish Pandey

et al.

Cluster Computing, Journal Year: 2025, Volume and Issue: 28(5)

Published: April 28, 2025

Language: Английский

Citations

0
Privacy-Preserving Federated Learning for Intrusion Detection in IoT Environments: A Survey DOI Creative Commons
Abhishek Vyas, Po‐Ching Lin, Ren‐Hung Hwang

et al.

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 127018 - 127050

Published: Jan. 1, 2024

Language: Английский

Citations

3
Privacy-Enhanced Sentiment Analysis in Mental Health: Federated Learning with Data Obfuscation and Bidirectional Encoder Representations from Transformers DOI Open Access
Shakil Ibne Ahsan, Djamel Djenouri,

Rehan Haider

et al.

Electronics, Journal Year: 2024, Volume and Issue: 13(23), P. 4650 - 4650

Published: Nov. 25, 2024

This research aims to find an optimal balance between privacy and performance in forecasting mental health sentiment. paper investigates federated learning (FL) augmented with a novel data obfuscation (DO) technique, where synthetic is used "mask" real points. Bidirectional Encoder Representations from Transformer (BERT) for sentiment analysis, forming new framework, FL-BERT+DO, that addresses the privacy-performance trade-off. With FL, remains decentralized, ensuring user-sensitive information retained on local devices rather than being shared FL server. The integration of BERT gives our system enhanced feature context sense-making text conduct, model extremely proficient emotion categorization tasks. experiments were performed combined (real replica synthetic) datasets containing emotions showed significant enhancements compared baseline methods. proposed FL-BERT+DO framework shows following metrics: prediction accuracy, 82.74%; precision, 83.30%; recall, F1-score, 82.80%. Further, we assessed its adversarial setup using membership inference linkage attacks ensure privacy-preserved did not suffer deeply. It demonstrates that, even large datasets, providing privacy-preserving possible can significantly improve existing methods addressing personal issues, like support. Based results work, propose development secure decentralized systems are capable high accuracy analysis meeting strict constraints.

Language: Английский

Citations

0
Enhancing Federated Learning Security: Combating Clients’ Data Poisoning with Classifier Ensembles DOI
Arunava Roy,

Dipankar Dasgupta

Published: Oct. 28, 2024

Language: Английский

Citations

0