Intelligent Malware Detection: Harnessing J48 Decision Trees and Gradient Boosting for Enhanced Security

Handling malware is crucial to application and system safety; yet, safeguarding against complex malware, such as metamorphic malware, presents an enormous challenge due to its ability to alter its structure and codes following each attack. Therefore, by classifying the executables and examining the presence of opcodes (functions), we provide a unique method in this research to identify complex malware with high accuracy. On the basis of discovered interesting characteristics, we examined the achievement of 13 classifiers using N-fold cross-validation accessible in machine learning (ML) program. In the group of these 13 classifiers we examined thoroughly based on hybrid model (The Gradient Boosting (GB) and J48). In these hybrid model, our methodology achieved an accuracy for detection of around 99.21% using the GB and J48 algorithm.