基于敏感权限和API的Android恶意软件家族分类方法

Android Malware Family Classification Method Based on Sensitive Permissions and API

提出一种基于敏感权限和API的Android恶意软件家族分类方法,通过提取敏感权限和敏感API,将两部分特征进行融合,构建特征库,最后结合随机森林算法进行恶意软件的家族分类。实验结果表明,该方法的检测精确度达到98.4%,显著优于其他基线算法,能够反映恶意软件的相似性和同源性。

A method of Android malware family classification based on sensitive permissions and APIs was proposed.After extracting sensitive permissions and sensitive APIs,the two features were fused to build a feature database.Finally,a random forest algorithm was used to classify malware families.Experimental results showed that the detection accuracy of this method reached 98.40%,which was significantly better than other baseline algorithms,and both the similarity and homology of malware were reflected.