Mobile Internet Malicious Application Detection Method Based on Support Vector Machine

Abstract

This paper demonstrates on the problem of detecting malicious applications in the mobile Internet, which is of great importance for personal information security and privacy security. Firstly, we obtain programs permissions and API call information through the reverse analysis of the APK file. Furthermore, we analyze the APK file directory structure and describe how to construct the feature vector for Android malicious applications detection. Secondly, we convert the mobile Internet malicious application detection problem to a classification problem, and utilize the SVM classifier to solve it. Finally, we conduct an experiment to test the performance of the proposed method. Experimental results that the proposed can detect mobile Internet malicious application with higher accuracy, true positive rate, and lower false positive rate.