基于决策树的Android恶意应用检测方法的研究

Android malware detection method based on decision tree

Android恶意应用对人们的危害越来越严重,因此,提出一种行之有效的恶意应用检测方法迫在眉睫。在Android系统中,一些敏感Ap I的操作都需要申请相应的权限,要实现某个恶意功能需要多个权限的配合。文中提取了应用程序的权限信息,利用频繁模式挖掘算法Apriori和决策树实现了对应用的检测分类。实验结果表明,该方法可行有效,检测率达90%。

The Android malware for our threaten is becoming more and more serious,so,it is imminent for us to put forward an effective malicious application detection method. In the Android system,some sensitive Ap I operations need to apply for the appropriate permissions. In order to achieve a malicious function,it requires multiple permissions to cooperate. This paper extracted the permissions of the application. In addition,this method achieved the detection classification of applications by Apriori and decision tree. The experimental results show that the method is effective and the detection rate is 90%.