Malware Classification Using Machine Learning

Cybersecurity is increasingly important in an era where technology is prevalent and vulnerable devices are integral to daily life. With the advent of new technologies such as artificial intelligence (AI), evolving cyber threats require innovative and dynamic solutions. One of these solutions is the automatic classification of malware within a system using AI, deep learning (DL), and machine learning (ML). In this paper, it is proposed to improve the reliability of malware detection through a modified multi-agent solution for the automatic classification of malware. The Malimg dataset consisting of twenty-five different classes of malware that have been turned into images is used. The proposed cascaded DL model represents an advancement over previous models on the same dataset, achieving a 97.7% accuracy.