Abstract
Android has become the most popular mobile operating system in the past ten years due to its three main advantages, namely, the openness of source code, richness of hardware selection, and millions of applications (apps). It is of no surprise that Android has become the major target of malware. The rapid increase in the number of Android malware poses big threats to smart phone users such as financial charges, information collection, and remote control. Thus, the in-depth study of the security issues of mobile apps is of great importance to the sound development of the smart phone ecosystem. We first introduce the existing problems and challenges of malware analysis, and then summarize the widely-used benchmark datasets. After that, we divide the existing malware analysis methods into three categories, including signature-based methods, machine learning-based methods, and behavior-based methods. We further summarize the techniques used in each method, and compare and analyze the advantages and disadvantages of different techniques. Finally, combined with our own research foundation in malware analysis, we explore and discuss future research directions and challenges.
Translated title of the contribution | Android malware detection: a survey |
---|---|
Original language | Chinese (Simplified) |
Pages (from-to) | 1148-1177 |
Number of pages | 30 |
Journal | Scientia Sinica Informationis |
Volume | 50 |
Issue number | 8 |
DOIs | |
Publication status | Published - 1 Aug 2020 |
Keywords
- Android
- Familial identification
- Machine learning
- Malware detection
ASJC Scopus subject areas
- General Computer Science
- Engineering (miscellaneous)