A Systematic Review on Hybrid Analysis using Machine Learning for Android Malware Detection

[PDF]

Abstract

Nowadays Android is the world’s most popular mobile operating system. Its pervasiveness also provokes the enormous growth of Android malware. Using machine learning methods to detect Android malware, researchers have focused on static analysis and dynamic analysis for most. But, different evasion techniques by shrewd malware authors made those techniques inadequate and ineffective. Therefore, recent researchers have turned their attention to the discovery of an effective strategy to combat. Hybrid analysis which is a fusion of static analysis and dynamic analysis would be a good candidate for that as it prevails over the individual shortcomings of static and dynamic analysis with the cost of complexity. Hybrid analysis has many opportunities as well as challenges. This research is intended to offer a detailed and systematic review of hybrid analysis using machine learning techniques for malware detection in Android. It encompasses leading hybrid analysis research: their contributions, strengths, and weaknesses. This work also discusses the challenges, opportunities, and future directions of hybrid analysis in detecting Android malware.

Keywords:

Hybrid Analysis, Android Malware Detection, Machine Learning