Innovative Malware Detection: Practical Swarm Optimization and fuzzyKNN Model in Honeypot Environment
Heba Othman
Abstract
Effective malware detection remains a critical challenge in cybersecurity. In this study, we propose an innovative method that combines swarm intelligence through Particle Swarm Optimization (PSO) with the fuzzy logic of the fuzzyKNN model, resulting in an adaptive and efficient malware detection system. Utilization of PSO assists in the selection of an optimal feature set from the malware dataset improving the performance of the fuzzyKNN model. create a secure and controlled environment for collecting diverse malware samples, we employ a honeypot. This controlled setting allows us to train our model without posing any risks to real operational systems. Conducted extensive tests to evaluate the effectiveness of our proposed methodology, comparing it against standard detection techniques. Our findings demonstrate the PSO-fuzzyKNN approach significantly enhances the accuracy of malware detection, outperforming traditional methods.contributes to advancement of malware detection technologies, offering a robust solution for addressing the evolving challenges posed by malicious software.