Performance Comparison of Whale and Harris Hawks Optimizers with Network Intrusion Prevention Systems

Abstract

Digital technology has permeated every aspect of our daily lives. Processing and evaluating information are highly demanding in all fields, including cybersecurity. Cybersecurity engineers widely use the Network Intrusion Prevention System (NIPS) to safeguard against cyberattacks. To avoid cyberattacks, the NIPS must deal with a large amount of data, which degrades its performance. This paper uses the whale optimization algorithm (WOA) and the Harris Hawks optimization method (HHO) to diminish the large amount of data that the NIPS needs to deal with. Subsequently, the Gradient Boosting Machine (GBM) is employed to determine the accuracy achieved when employing WOA and HHO. The GBM classifier is widely regarded as a sophisticated and straightforward classifier in data mining. Regardless of the premise of feature independence, it outperforms all other classification algorithms by delivering excellent performance. When using GBM, the findings indicate that the accuracy achieved with HHO is 89.81%, but the accuracy attained with WOA is 94.3%.