期刊名称：IAENG International Journal of Computer Science
印刷版ISSN：1819-656X
电子版ISSN：1819-9224
出版年度：2018
卷号：45
期号：4
页码：514-522
出版社：IAENG - International Association of Engineers
摘要：In this paper, a machine learning based techniqueis proposed to classify android applications in three classesbased on the confidence level defined as safe, suspicious andhighly suspicious. Thirty six features are extracted and selectedfrom Mobile Security Framework based on penetration testing.A set of experiments has been conducted on the scale of 13,850android applications which includes 8,782 android applicationsdownloaded from apk-dl.com, 3,960 malware and 1,108 benignapplications. In order to compare the accuracy of the classificationmodel, a ground truth of the confidence level is created byusing VirusTotal. The proposed method can detect and classifyandroid applications into three confidence levels with 81.80%accuracy. Experiment for binary classification, classify as beingmalware or benign has yielded 93.63% accuracy.
