Enhancing Intrusion Detection Systems: A Comparative Study using Whale Optimization Algorithm and Hawks Optimization Algorithm
Mosleh M. Abualhaj, Sumaya Nabil Al-Khatib, Ali Al-Allawee, Alhamza Munther, Mohammed Anbar · 2024
Intrusion Detection Systems (IDS) play a pivotal role in safeguarding computer networks against malicious activities. This research explores the efficacy of two optimization algorithms, namely Whale Optimization Algorithm (WOA) and Hawks Optimization (HHO), for feature selection to enhance the performance of IDS. The NSL-KDD dataset is utilized as a benchmark to evaluate the impact of selected features on classification accuracy. The study employs the Random Forest (RF) classification algorithm to evaluate the effectiveness of feature selection methods. Implementation is conducted in Python, leveraging its versatile libraries and frameworks. Comparative analysis reveals that when integrating the selected features, RF attains superior results compared to utilizing features separately for binary and multiclass classifications. Results indicate that RF with HHO achieves an impressive accuracy of 98.02%, showcasing the prowess of HHO in enhancing IDS performance. Meanwhile, RF with WOA achieves a commendable accuracy of 97.81%.