Android malware analysis approach based on control flow graphs and machine learning algorithms

• Published in: 2016 4th International Symposium on Digital Forensic and Security (ISDFS) pp. 26 - 31
• Main Authors: Atici, Mehmet Ali, Sagiroglu, Seref, Dogru, Ibrahim Alper
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.04.2016
• Subjects: Android; control flow graphs; Decision support systems; Flow graphs; machine learning; Malware; Mobile communication; mobile security; Security; static analysis
• DOI: 10.1109/ISDFS.2016.7473512

Smart devices from smartphones to wearable computers today have been used in many purposes. These devices run various mobile operating systems like Android, iOS, Symbian, Windows Mobile, etc. Since the mobile devices are widely used and contain personal information, they are subject to security attacks by mobile malware applications. In this work we propose a new approach based on control flow graphs and machine learning algorithms for static Android malware analysis. Experimental results have shown that the proposed approach achieves a high classification accuracy of 96.26% in general and high detection rate of 99.15% for DroidKungfu malware families which are very harmful and difficult to detect because of encrypting the root exploits, by reducing data dimension significantly for real time analysis.