Enhancing Intrusion Detection in Software Defined Networks with Optimized Feature Selection and Logistic Regression

In this study, we present a highly effective machine learning model for intrusion detection in Software Defined Networks (SDN), showcasing remarkable accuracy and precision in identifying network threats. Our approach utilizes an extensive dataset, covering a wide array of network flow statistics to differentiate between normal and malicious traffic. The model's robustness is demonstrated through an accuracy of 98%, with precision and recall metrics substantiated by F1-scores near 0.98. This research not only addresses the intricacies of SDN environments but also offers a scalable solution for evolving cyber-security challenges. Our findings mark a significant advancement in network security, providing a comprehensive framework for future developments in the field of intrusion detection systems.