MAPS: a dataset for semantic profiling and analysis of Android applications

MAPS: a dataset for semantic profiling and analysis of Android applications

연구 분야: Safety

학회: MobiArch '22: Proceedings of the 17th ACM Workshop on Mobility in the Evolving Internet Architecture

The vulnerability of smartphones to cyber attacks has been a serious concern for users emerging from the integrity of installed mobile applications (apps). These applications are fundamentally developed to provide legitimate and diversified on-the-go services; however, harmful and dangerous ones have found the perfect door to get into smartphones performing malicious behaviors. More effective and sophisticated application analysis is a practical strategy to reveal malicious activities and provide more insights into the application behavior. In this paper, we present the Malware Analysis and Profiling on Smartphones (MAPS) dataset for the research community that provides a very in-depth and detailed deep analysis of applications for extracting an enhanced set of features. The MAPS dataset relies on a diverse dataset collected from different repositories, containing more than 153000 applications (>2TB in size), and outputs more than 40000 features for analysis and training of deep neural network models.