文章基本信息

标题：Permission Extraction Framework for Android Malware Detection
其他标题：Permission Extraction Framework for Android Malware Detection
本地全文：下载
作者：Ali Ghasempour ; Nor Fazlida Mohd Sani ; Ovye John Abari 等
期刊名称：International Journal of Advanced Computer Science and Applications(IJACSA)
印刷版ISSN：2158-107X
电子版ISSN：2156-5570
出版年度：2020
卷号：11
期号：11
DOI：10.14569/IJACSA.2020.0111159
出版社：Science and Information Society (SAI)
摘要：Nowadays, Android-based devices are more utilized than other Operating Systems based devices. Statistics show that the market share for android on mobile devices in March 2018 is 84.8 percent as compared with only 15.1 percent iOS. These numbers indicate that most of the attacks are subjected to Android devices. In addition, most people are keeping their confidential information on their mobile phones, and hence there is a need to secure this operating system against harmful attacks. Detecting malicious applications in the Android market is becoming a very complex procedure. This is because as the attacks are increasing, the complexity of feature selection and classification techniques are growing. There are a lot of solutions on how to detect malicious applications on the Android platform but these solutions are inefficient to handle the features extraction and classification due to many false alarms. In this work, the researchers proposed a multi-level permission extraction framework for malware detection in an Android device. The framework uses a permission extraction approach to label malicious applications by analyzing permissions and it is capable of handling a large number of applications while keeping the performance metrics optimized. A static analysis method was employed in this work. Support Vector Machine (SVM) and Decision Tree Algorithm was used for the classification. The results show that while increasing input data, the model tries to keep detection accuracy at an acceptable level.
关键词：Malware detection; android device; operating system; malicious application; machine learning