Performance Evaluation of Whale and Harris Hawks Optimization Algorithms with Intrusion Prevention Systems

Intrusion Prevention Systems (IPS) protect computer networks from hostile activities. This study investigates the effectiveness of two optimization algorithms, Hawks Optimization (HHO) and Whale Optimization Algorithm (WOA), in selecting features to improve the performance of Intrusion Prevention Systems (IPS). The NSL-KDD dataset is employed as a standard to assess the influence of specific features on classification accuracy. The study uses the K-nearest neighbors (KNN) classification algorithm to assess the efficacy of feature selection techniques. The implementation is carried out using Python, using its flexible libraries and frameworks. Comparative investigation shows that when combining the chosen features, KNN achieves better outcomes than using the whole dataset of features. The results demonstrate that the combination of KNN with WOA provides a remarkable accuracy rate of 97.42%, highlighting the effectiveness of WOA in improving IPS performance. On the other hand, the combination of KNN and HHO demonstrates an impressive accuracy rate of 96.87%.