Windows Malware Detection Under the Machine Learning Models and Neutrosophic Numbers

Significant cybersecurity risks are posed by malware assaults on Windows computers, which call for efficient detection and prevention systems. Supervised machine learning classifiers have shown great promise in the field of malware detection. Comprehensive research comparing the effectiveness of various classifiers, particularly for Windows malware detection, is still required. Closing this gap can yield valuable information for improving cybersecurity tactics. A thorough comparison of supervised classifiers for Windows malware detection is lacking, even though several research have investigated malware detection using machine learning approaches. Determining the relative efficacy of these classifiers can help choose the best detection techniques and enhance security protocols in general. This study applies to 6 ML models for Windows malware detection. After that, we evaluate these models using the neutrosophic set to overcome the uncertainty information. The single values neutrosophic sets (SVNSs) are used in this study. The EDAS method is used to select the based model under the evaluation matrices.