Enhanced DDoS Attack Detection in IoT Environments Using Voting and Stacking Ensemble Learning: Implementation and Performance Analysis
Year: 2025
Type: article
Abstract: Due to the increase in the adoption of Internet of Things (IoT) devices, there has been a significant increase in Distributed Denial of Service (DDoS) attack. This is because IoT devices have introduced significant security vulnerabilities, thereby increasing the attack surface. This paper aims to present an enhanced DDoS Attack detection in an IoT environment using an ensemble learning approach. This was achieved by implementing the voting and stacking classifiers that combine four supervised learning algorithms: Random Forest, Decision Trees, Logistic Regression, and K-Nearest Neighbors. Using the comprehensive CIC-IoT2023 dataset, the results of the test conducted indicate outstanding performance, with the voting classifier achieving 99.39% accuracy (190.9872ms inference time, 32 false positives) and the stacking classifier reaching 99.40% accuracy (224.9587ms inference time, 89 false positives), a 5-fold stratified cross-validation was conducted which validated the models' robustness as a significant improvement on previous study conducted in this area.
Source: UMYU Scientifica
Authors: Abdulrahman Tunde Alabelewe, Muhammad Aminu Ahmad, Ahmed Abubakar Aliyu, Mohammed Ibrahim, Abubakar Muazu Ahmed, Saadatu Abdulkadir
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4416715365