Detection of Android Malware using Feature Selection with a Hybrid Genetic Algorithm and Simulated Annealing (SVM and DBN)

Abstract

Because of the widespread use of the Android operating system and the simplicity with which applications can be created on the Android platform, anyone can easily create malware using pre-made tools. Due to the spread of malware among many helpful applications, Android users are experiencing issues. In this study, we showed how to use permissions gleaned from static analysis to identify Android malware. Utilising support vector machines and deep belief networks, we choose the pertinent features from the set of permissions based on this methodology. The suggested technique increases the effectiveness of Android malware detection.