An Intrusion Detection System for SDN Using Machine Learning

Software Defined Networking (SDN) has emerged as a promising and exciting option for the future growth of internet. SDN increased flexibility transparency managed, centralized, controlled network. On other hand, these advantages create more vulnerable environment with substantial risks, culminating in network difficulties, system paralysis, online banking frauds, robberies. These issues have significant detrimental impact on organizations, enterprises, even economies. Accuracy, high performance, real-time systems are necessary to achieve this goal. Using extend intelligent machine learning methodologies an Intrusion Detection System (IDS) stimulated interest numerous research investigators over last decade. In paper, novel HFS-LGBM IDS is proposed SDN. First, Hybrid Feature Selection algorithm consisting two phases applied reduce data dimension obtain optimal feature subset. first phase, Correlation based (CFS) used The set obtained by applying Random Forest Recursive Elimination (RF-RFE) second phase. A LightGBM then detect classify different types attacks. experimental results NSL-KDD dataset show that produces outstanding compared existing methods terms accuracy, precision, recall f-measure.