Evaluation of Federated Learning for Robust IoT Security Against Label Flipping Attacks

Maryam Alsereidi; Zeyar Aung; Panos Liatsis

doi:10.1007/978-981-95-2680-2_43

Evaluation of Federated Learning for Robust IoT Security Against Label Flipping Attacks

Maryam Alsereidi
, Zeyar Aung
, Panos Liatsis

Center for Cyber-Physical Systems

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

The increase of the Internet of Things (IoT) ecosystem demands robust security mechanisms to address emerging cyber threats. This paper evaluates the performance of Machine Learning (ML) and Federated Learning (FL) approaches on the BoTNeTIoT-L01 dataset, which represents network traffic from various IoT devices under cyberattack scenarios. Findings demonstrate that while traditional ML models provide high accuracy, reaching up to 99.99%, FL approaches excel in preserving privacy and maintaining robust performance metrics. Even under adversarial conditions such as data poisoning attacks, FL methods show resilience, with accuracy levels sustaining above 90% in non-extreme cases. This study highlights the potential of FL in advancing secure, scalable frameworks for IoT security.

Original language	British English
Title of host publication	Synergies in Data Analytics and Cyber Security - Proceedings of the International Conference, DACS 2024
Editors	Deepak Puthal, Bijaya Ketan Panigrahi, Niranjan Ray, Zhiguo Ding
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	553-569
Number of pages	17
ISBN (Print)	9789819526796
DOIs	https://doi.org/10.1007/978-981-95-2680-2_43
State	Published - 2026
Event	7th International Conference on Data Analytics and Cyber Security, DACS 2024 - Bodh Gaya, India Duration: 20 Dec 2024 → 22 Dec 2024

Publication series

Name	Lecture Notes in Electrical Engineering
Volume	1479 LNEE
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	7th International Conference on Data Analytics and Cyber Security, DACS 2024
Country/Territory	India
City	Bodh Gaya
Period	20/12/24 → 22/12/24

Keywords

Cyber threats
Data poisoning
Federated learning
Intrusion detection systems
Iot security
Machine learning

Access to Document

10.1007/978-981-95-2680-2_43

OpenUrl availability

Full text

Cite this

Alsereidi, M., Aung, Z., & Liatsis, P. (2026). Evaluation of Federated Learning for Robust IoT Security Against Label Flipping Attacks. In D. Puthal, B. K. Panigrahi, N. Ray, & Z. Ding (Eds.), Synergies in Data Analytics and Cyber Security - Proceedings of the International Conference, DACS 2024 (pp. 553-569). (Lecture Notes in Electrical Engineering; Vol. 1479 LNEE). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-95-2680-2_43

Alsereidi, Maryam ; Aung, Zeyar ; Liatsis, Panos. / Evaluation of Federated Learning for Robust IoT Security Against Label Flipping Attacks. Synergies in Data Analytics and Cyber Security - Proceedings of the International Conference, DACS 2024. editor / Deepak Puthal ; Bijaya Ketan Panigrahi ; Niranjan Ray ; Zhiguo Ding. Springer Science and Business Media Deutschland GmbH, 2026. pp. 553-569 (Lecture Notes in Electrical Engineering).

@inproceedings{e39626a606f943439eb1704ef2c15251,

title = "Evaluation of Federated Learning for Robust IoT Security Against Label Flipping Attacks",

abstract = "The increase of the Internet of Things (IoT) ecosystem demands robust security mechanisms to address emerging cyber threats. This paper evaluates the performance of Machine Learning (ML) and Federated Learning (FL) approaches on the BoTNeTIoT-L01 dataset, which represents network traffic from various IoT devices under cyberattack scenarios. Findings demonstrate that while traditional ML models provide high accuracy, reaching up to 99.99\%, FL approaches excel in preserving privacy and maintaining robust performance metrics. Even under adversarial conditions such as data poisoning attacks, FL methods show resilience, with accuracy levels sustaining above 90\% in non-extreme cases. This study highlights the potential of FL in advancing secure, scalable frameworks for IoT security.",

keywords = "Cyber threats, Data poisoning, Federated learning, Intrusion detection systems, Iot security, Machine learning",

author = "Maryam Alsereidi and Zeyar Aung and Panos Liatsis",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.; 7th International Conference on Data Analytics and Cyber Security, DACS 2024 ; Conference date: 20-12-2024 Through 22-12-2024",

year = "2026",

doi = "10.1007/978-981-95-2680-2\_43",

language = "British English",

isbn = "9789819526796",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "553--569",

editor = "Deepak Puthal and Panigrahi, \{Bijaya Ketan\} and Niranjan Ray and Zhiguo Ding",

booktitle = "Synergies in Data Analytics and Cyber Security - Proceedings of the International Conference, DACS 2024",

address = "Germany",

}

Alsereidi, M, Aung, Z & Liatsis, P 2026, Evaluation of Federated Learning for Robust IoT Security Against Label Flipping Attacks. in D Puthal, BK Panigrahi, N Ray & Z Ding (eds), Synergies in Data Analytics and Cyber Security - Proceedings of the International Conference, DACS 2024. Lecture Notes in Electrical Engineering, vol. 1479 LNEE, Springer Science and Business Media Deutschland GmbH, pp. 553-569, 7th International Conference on Data Analytics and Cyber Security, DACS 2024, Bodh Gaya, India, 20/12/24. https://doi.org/10.1007/978-981-95-2680-2_43

Evaluation of Federated Learning for Robust IoT Security Against Label Flipping Attacks. / Alsereidi, Maryam; Aung, Zeyar ; Liatsis, Panos.
Synergies in Data Analytics and Cyber Security - Proceedings of the International Conference, DACS 2024. ed. / Deepak Puthal; Bijaya Ketan Panigrahi; Niranjan Ray; Zhiguo Ding. Springer Science and Business Media Deutschland GmbH, 2026. p. 553-569 (Lecture Notes in Electrical Engineering; Vol. 1479 LNEE).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Evaluation of Federated Learning for Robust IoT Security Against Label Flipping Attacks

AU - Alsereidi, Maryam

AU - Aung, Zeyar

AU - Liatsis, Panos

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.

PY - 2026

Y1 - 2026

N2 - The increase of the Internet of Things (IoT) ecosystem demands robust security mechanisms to address emerging cyber threats. This paper evaluates the performance of Machine Learning (ML) and Federated Learning (FL) approaches on the BoTNeTIoT-L01 dataset, which represents network traffic from various IoT devices under cyberattack scenarios. Findings demonstrate that while traditional ML models provide high accuracy, reaching up to 99.99%, FL approaches excel in preserving privacy and maintaining robust performance metrics. Even under adversarial conditions such as data poisoning attacks, FL methods show resilience, with accuracy levels sustaining above 90% in non-extreme cases. This study highlights the potential of FL in advancing secure, scalable frameworks for IoT security.

AB - The increase of the Internet of Things (IoT) ecosystem demands robust security mechanisms to address emerging cyber threats. This paper evaluates the performance of Machine Learning (ML) and Federated Learning (FL) approaches on the BoTNeTIoT-L01 dataset, which represents network traffic from various IoT devices under cyberattack scenarios. Findings demonstrate that while traditional ML models provide high accuracy, reaching up to 99.99%, FL approaches excel in preserving privacy and maintaining robust performance metrics. Even under adversarial conditions such as data poisoning attacks, FL methods show resilience, with accuracy levels sustaining above 90% in non-extreme cases. This study highlights the potential of FL in advancing secure, scalable frameworks for IoT security.

KW - Cyber threats

KW - Data poisoning

KW - Federated learning

KW - Intrusion detection systems

KW - Iot security

KW - Machine learning

UR - https://www.scopus.com/pages/publications/105030206492

U2 - 10.1007/978-981-95-2680-2_43

DO - 10.1007/978-981-95-2680-2_43

M3 - Conference contribution

AN - SCOPUS:105030206492

SN - 9789819526796

T3 - Lecture Notes in Electrical Engineering

SP - 553

EP - 569

BT - Synergies in Data Analytics and Cyber Security - Proceedings of the International Conference, DACS 2024

A2 - Puthal, Deepak

A2 - Panigrahi, Bijaya Ketan

A2 - Ray, Niranjan

A2 - Ding, Zhiguo

PB - Springer Science and Business Media Deutschland GmbH

T2 - 7th International Conference on Data Analytics and Cyber Security, DACS 2024

Y2 - 20 December 2024 through 22 December 2024

ER -

Alsereidi M, Aung Z , Liatsis P. Evaluation of Federated Learning for Robust IoT Security Against Label Flipping Attacks. In Puthal D, Panigrahi BK, Ray N, Ding Z, editors, Synergies in Data Analytics and Cyber Security - Proceedings of the International Conference, DACS 2024. Springer Science and Business Media Deutschland GmbH. 2026. p. 553-569. (Lecture Notes in Electrical Engineering). doi: 10.1007/978-981-95-2680-2_43