Enhancing Malware Detection with Firefly and Grey Wolf Optimization Algorithms
Ahmad Adel Abu-Shareha, Mosleh M. Abualhaj, Ali Al-Allawee, Alhamza Munther, Mohammed Anbar
Abstract
Malware poses a serious threat to computer networks, user privacy, and user systems. Effective cybersecurity depends on the correct detection and classification of malware. In order to improve the effectiveness of malware detection systems, this research paper explores the application of two optimization algorithms, Firefly Algorithm (FA) and Grey Wolf Optimization (GWO), for feature selection in the context of malware detection using the CIC-MaIMem-2022 dataset. Leveraging Python for implementation and employing Random Forest (RF) as the clas-sification model, the study compares the performance of FA and GWO in enhancing malware detection accuracy. Results indicate that the RF model with FA achieves remarkable metrics, boasting an accuracy, precision, recall, and Fl-score of 99.97%, along with a Matthews Correlation Coefficient (MCC) of 99.95%. In contrast, the GWO-based approach yields an accuracy, precision, recall, and Fl-score of 99.95%, accompanied by an MCC of 96.88%. The findings highlight the efficacy of FA over GWO in optimizing feature selection for enhancing malware detection accuracy, providing valuable insights for future research in cybersecurity..