A Hybrid Malware Detection System for Enhanced Cloud Security Utilizing Trust-Based Glow-Worm Swarm Optimization and Recurrent Deep Neural Networks

User credentials are vulnerable to exposure in demilitarized zones due to software vulnerabilities and hardware threats. This research aims to mitigate these risks by proposing a sophisticated trust-based malware detection (T-MALWARE DETECTION) method that can accurately classify data. The proposed system utilizes an enhanced Glow-Worm Swarm Optimization (IGWSO) technique to efficiently cluster datasets. To classify potential intrusions and assign trust levels to cloud data after clustering, a Recurrent Neural Network (RNN) approach is employed. The effectiveness of the Trust-oriented Malware Detection System (T-MALWARE DETECTIONS) is evaluated using metrics such as detection rate, precision, recall, and F-measure. This system is developed using Java and the CloudSimulator (CloudSim) tool, allowing for a thorough evaluation of its performance in comparison to contemporary state-of-the-art systems.