基于类别以及权限的Android恶意程序检测

Detection of malicious Android application based on permissions and category

针对Android平台恶意软件数量的日益增多,提出一种基于类别以及权限的Android恶意程序检测方法。以Google Play划分的类为依据,统计每一个类别应用程序权限使用情况,利用应用程序的访问权限,计算该类别恶意阈值。安装应用程序时,利用序列最小优化算法给应用程序正确分类,分析应用程序使用的权限,计算该程序恶意值,与该类别的恶意阈值进行比较,给用户提供建议,帮助用户判断该程序是否是恶意的。实验结果表明了该方法的有效性和可行性。

To address the increasement of malware for Android, based on different categories and permissions of Android applications malware detection method is constructed. The method is depending on the categories in the Google Play as a reference to access permissions of applications of each category. The permissions of applications is analyzed and the malicious threshold of this category is calculated. When the application is installed, sequential minimal optimization algorithm is used to classify the application, and the permissions of the application is analyzed and the malicious value of this application is calculated. Comparing the malicious value and the category＇s malicious threshold, users are advised to determine the program＇s security. Finally, the effectiveness and feasibility of the method is verified experimentally.