Issue |
ITM Web Conf.
Volume 70, 2025
2024 2nd International Conference on Data Science, Advanced Algorithm and Intelligent Computing (DAI 2024)
|
|
---|---|---|
Article Number | 01015 | |
Number of page(s) | 8 | |
Section | Traffic Prediction and Analysis | |
DOI | https://doi.org/10.1051/itmconf/20257001015 | |
Published online | 23 January 2025 |
Machine Learning-Based Network Detection Research for SDNs
School of Computer and Cyberspace Security, Fujian Normal University, Fuzhou city, China
Corresponding author: 121152022048@student.fjnu.edu.cn
This research endeavors to fortify the security posture of Software-Defined Networks (SDN) through the strategic utilization of intelligent machine learning techniques, with a primary focus on mitigating detrimental Denial of Service (DoS) attacks. To accomplish this, this study constructed a rigorously designed simulated SDN environment, which served as the cornerstone for meticulously assembling a comprehensive dataset encompassing a diverse array of attack vectors, with particular emphasis on DoS. Employing a tactical blend of established and cutting-edge machine learning algorithms, including Random Forest, Logistic Regression, and Decision Tree, alongside the advanced XGBoost and LightGBM models, this study conducted an exhaustive investigation to pinpoint the most efficacious methods for swiftly and precisely identifying DoS threats. It is necessary to note that XGBoost and LightGBM demonstrate an astonishing level of multiple performance, which testifies their outstanding ability to enhance SDN security. Reasserting the idea of the critically important role of machine learning for securing SDNs against possible intrusions, these results point not only to the highly beneficial applications of machine learning for protecting SDNs against malicious intrusions but also its indispensable role in preserving network stability and optimizing performance. Moreover, it emphasizes the operational advantage of deploying multiple organic sets of machine learning algorithms, which can achieve even greater precision and efficiency than individual machine learning algorithms in practical uses, bringing it closer to developing a more robust and secure SDN environment.
© The Authors, published by EDP Sciences, 2025
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.