Cyber Threat Detection in Software-Defined Networks: An Empirical Analysis of Machine Learning Methods
Ishu Sharma, Jiya Saini, Gunjan Chhabra, Keshav Kaushik · 2023
The field of cyber threat detection in Software-Defined Networks (SDNs) is explored in this research article, which provides a thorough model supported by machine learning techniques. In order to strengthen network security, the model installs AI-Enabled Servers at the application and control levels. These servers are designed to quickly detect and neutralize any threats. These servers are equipped with six different classifiers that enable them to make well-informed decisions based on a range of network metrics: Random Forest, K-Nearest Neighbor, Support Vector Machine, Naïve Bayes, Decision Tree, and Logistic Regression. The first line of protection in the model is provided by an initial firewall, which is followed by AI-Enabled Servers for quick threat detection. The research work includes empirical analysis that demonstrates the effectiveness of the model and suggests future directions for practical application, improved threat detection, adaptable security policies, ethical considerations, resource optimization, and industry expert collaboration—all of which will further the search for reliable SDN security solutions. Empirical investigation has confirmed the robustness of the proposed model and shown its effectiveness in detecting and mitigating cyber risks in SDN systems. AI-Enabled Servers are able to make well-informed decisions and adjust to changing network conditions by being trained on a variety of network metrics, such as Port Duration Time, Port Bytes, Port Packets, RX Errors, TX Errors, Port Collisions, Table Active Count, Table Lookup Count, and Table Matched Count.