A Survey on Techniques for Android Malware Detection

Abstract

Android is an open-source platform for numerous applications. It is playing a major role in the current world and used for handling users personal and confidential data. Android mobile users can download various applications and certainly upload them easily without any cost and authorization through Google play store. Due to this, Android threats may occur, and these are spreading easily leading to various types of Android malwares which are growing around the planet and effecting user’s personal data, their systems, and reputed organizations harming the sensitive information. There are two mechanisms which propagate for Android malware detection. They are signature-based techniques and permission-based techniques. Signature-based techniques were used for detection of unknown malwares based on signature samples, but they cannot detect newly discovered threats like zero-day attacks, which are not known to the world before they are seen in the wild. This paper deals with a survey on malware detection techniques. In this survey, it is observed that Android malware detection techniques of permission-based systems perform more efficiently. There are discussions on various detection mechanisms and machine learning algorithms used in Android malware detection, and this paper highlights their advantages as well as disadvantages.