User-Based Solutions for Android Malware Detection

The existence of Android malware is increasing day by day. These software are able to access the information of users in the background without being aware of the users on smart devices, send SMS to the user name and violate the privacy of users. There are different methods for detecting malware such as static, dynamic and hybrid. However, each method has its own specific shortcomings. In this paper, it is aimed to create a detection rule with dynamic methods. In this way, both the user and the user will be informed about the instant attacks. For this purpose, firstly, statistics about the usage of operating system methods in a device are examined and average value ranges were determined. Then, this rule is updated with the usage frequency of the users and a specific statistical rule is obtained for the user. The abnormally defined unregistered operating system movements on the mobile device are detected and the user is informed instantly.