An Ensemble approach for advance malware memory analysis using Image classification techniques

New types of malware have emerged due to the increased use of computer systems and web services, which are unsafe and harder to identify. The latest reports show that a new type of file-less malware infects users' systems but leaves no trace on the system's hard disk. The current static malware analysis techniques cannot detect malware that utilizes encryption and deception techniques. To detect and safeguard from this malware, in our study, ensemble-based machine learning approaches are implemented and optimized. The models are combined using different voting processes. The binary Windows malware and benign files are converted to image files and analyzed using popular learning techniques. This study profoundly analyses the images and classifies the classes into benign and malicious. The proposed ensemble approach achieves 97.17% accuracy as compared with other popular methods.