A Hybrid Malware Vulnerability using Deep Learning Technique

Abstract

Ransomware is on the rise and effective defensefrom it is of utmost importance to guarantee security of mobile users’ data. Current solutions provided by antimalware vendors are signature-based and thus ineffective in removing ransomware and restoring the infected devices and files. To detect ransomware, the hybrid approach was developed using modified random forest and deep learning technique. The performance of our hybrid detection method is evaluated on a dataset that contains both ransomware and legitimate applications. Additionally, the performance of the static and of the dynamic stand-alone methods for comparison is evaluated. Results showed that detection method perform well in detecting ransomware, their combination in a form of a hybrid method performs best, being able to detect ransomware with 98% precision and having a false positive rate of less than 4%.