Multi-Stage Optimized Machine Learning Framework for Network Intrusion Detection

Cyber-security garnered significant attention due to the increased dependency of individuals and organizations on Internet their concern about security privacy online activities. Several previous machine learning (ML)-based network intrusion detection systems (NIDSs) have been developed protect against malicious behavior. This paper proposes a novel multi-stage optimized ML-based NIDS framework that reduces computational complexity while maintaining its performance. work studies impact oversampling techniques models’ training sample size determines minimal suitable size. Furthermore, it compares between two feature selection techniques, information gain correlation-based, explores effect performance time complexity. Moreover, different ML hyper-parameter optimization are investigated enhance NIDS’s The proposed is evaluated using recent datasets, CICIDS 2017 UNSW-NB 2015 datasets. Experimental results show model significantly required (up 74%) set 50%). enhanced with accuracies over 99% for both outperforming literature works by 1-2% higher accuracy lower false alarm rate.