Title: Strengthening IoT security: assessing ensemble machine learning for cloud DDoS attack protection

Authors: Bijay Kumar Paikaray; Lalmohan Pattnaik

Addresses: Centre for Data Science, Department of Computer Science and Engineering, Siksha 'O' Anusandhan (Deemed to be University), Bhubaneswar, Odisha, 751030, India ' Faculty in Emerging Technologies, Sri Sri University, Cuttack, 754006, India

Abstract: The vulnerability to distributed denial of service (DDoS) attacks has significantly increased due to the simultaneous advancement of cloud services and the Internet of Things (IoT). This has facilitated the ability of unscrupulous individuals to interrupt cloud services and harm the reputation of organisations. Due to the unique characteristics and constraints of IoT devices, traditional approaches to identifying DDoS attacks can prove inadequate within an IoT setting. The performance of the five supervised learning models Logistic Regression (LR), Ridge Classifier (RC), AdaBoostClassifier (ADB), and ExtraTrees Classifier (ETC) are evaluated in the accurate identification of IoT-based network activities. The evaluation of the learning models is done on a subclass of CIC DoS, CICIDS-2017, and CSECICIDS-2018 datasets, with a special focus on 2017 data. The feature engineering approach is employed to improve the learning model's accuracy. The experimental results revealed the highest level of accuracy rate of 99.97% for ExtraTreeClassifier.

Keywords: DDoS; distributed denial of service; IoT; Internet of Things; machine learning; IDS; intrusion detection system.

DOI: 10.1504/IJCBDD.2025.151204

International Journal of Computational Biology and Drug Design, 2025 Vol.16 No.4, pp.350 - 365

Received: 15 Apr 2024
Accepted: 01 Jan 2025
Published online: 19 Jan 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article