With the rapid development of the mobile network, large numbers of Android malware emerge and pose a serious security threat to Android users. The function call graphs (FCG) extracted from Android application involves permissions, API calls, and structure semantics. Leveraging FCGs has great potential for Android application detection. In this paper, we propose AFCGDroid, an approach based on attributed function call graph (AFCG), to detect Android malware. The nodes of FCGs are divided into internal nodes and external nodes in our work. AFCGDroid extracts FCGs through static analysis and then generates AFCGs by labeling its internal nodes with the external nodes. To handle the graph data, graph embedding based on deep learning is proposed to embed the AFCGs into low dimension vectors. We evaluate AFCGDroid in malware detection achieving a 99.6% accuracy in a publicly available dataset, Drebin.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.