作者: Patrick Carter , Collin Mulliner , Martina Lindorfer , William Robertson , Engin Kirda
DOI: 10.1007/978-3-662-54970-4_13
关键词: Computer science 、 Mobile computing 、 Human–computer interaction 、 User interface 、 Code coverage 、 Current user 、 Android application 、 Android (operating system) 、 Mobile malware 、 Market share
摘要: Mobile computing has experienced enormous growth in market share and computational power recent years. As a result, mobile malware is becoming more sophisticated prevalent, leading to research into dynamic sandboxes as widespread approach for detecting malicious applications. However, the event-driven nature of Android applications renders critical capability automatically generate deterministic intelligent user interactions drive analysis subjects improve code coverage. In this paper, we present CuriousDroid, an automated system exercising application interfaces intelligent, user-like manner. CuriousDroid operates by decomposing on-the-fly creating context-based model that tailored current layout. We integrated with Andrubis, well-known sandbox, conducted large-scale evaluation 38,872 taken from different data sets. Our demonstrates significant improvements both end-to-end sample classification well increases raw number elicited behaviors at runtime.