ABSTRACT
User credentials are exposed or might end up in a demilitarized zone due to a variety of software vulnerabilities and hardware threats. The research project aims to investigate and ultimately suggest a trust-based malware detection (TMD) method for the best possible classification of data. An enhanced Glow-Worm Swarm Optimization (IGWSO) technique is suggested to organize the taken dataset into various clusters. Recurrent deep neural network-(RNN) based computing is utilized to classify the suspected intrusion and derive varied trust levels for the cloud data after clustering. Rate of detection, precision, recall, and F-measures are the metrics used by the introduced Trust oriented Malware Detection System (TMDS) system, which is built in Java using the Cloud Simulator (CloudSim) tool to determine the algorithm's efficacy over current state-of-the-art systems.
