Abstract

In light of the changing landscape of cybersecurity, accurate malware detection and identification remain relevant issues. The following paper presents a new malware detection model for optimizing detection efficiency and scalability with the Grey Wolf Optimizer (GWO) algorithm and Distributed Convolutional Neural Networks (CNNs). Apart from improving the scalability of the system, the suggested distributed architecture increases its adaptability to new and advanced malware risks. Experimental evaluation of the proposed framework using the Malimg benchmark dataset demonstrates substantial enhancements in key performance indicators such as accuracy, precision, recall, and F-score. Achieving a remarkable 99.95% accuracy, the framework underscores its reliability and practical applicability in real-world cybersecurity contexts. These results confirm the efficacy of metaheuristic optimization methods in combination with deep learning algorithms in solving the complexity of modern malware detection. This contribution enhances the state of the art in cybersecurity by providing a systematic and scalable solution to the ever-changing nature of malware threats.