A Hybrid Detection Method for Android Malware
Qi Fang, Xiaohui Yang, Ce Ji · 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC) · 2019
With the increasing market share of Android devices, malicious applications are developing and spreading rapidly. So it is imperative to improve the detection accuracy of Android malware. In this paper, we propose a hybrid detection method that performs dynamic detection on the results of static detection. The proposed method extracts the static features and dynamic features of the application, which can better detect the maliciousness of the Android application. Furthermore, we present experimental results of three ensemble methods in the dynamic detection, and choose the XGBoost algorithm with the optimum performance. Finally, we show that our method achieves a detection accuracy of 94.6%, which is higher than 85.3% of the static detection and 94.1% of the dynamic detection.