SELMAnalyzer: An APP Privacy Leakage Analysis Method Based on Program Execution Logic

Mobile phones have become an indispensable part of people's daily lives, but many mobile applications continue to obtain and use people's private data through malicious behaviors hidden behind normal functions. Existing work provides many methods to detect malicious applications' behaviors. However, the analysis and detection technology for the logic triggering process and implementation methods of potential malicious behaviors still need further improvement. This paper proposes an APP privacy leakage analysis method called SELMAnalyzer. This method builds a software execution logic analysis model SELM(Software Execution Logical Model) based on a graph structure to model the dynamic execution logic and implementation methods of Android applications and mine the privacy leakage problems of applications hidden behind normal UI function interfaces or normal components. At the same time, the method proposed in this paper is verified using a real Android application dataset with an accuracy rate of 89%. Compared with similar methods, it can intuitively display the complete calling process of privacy theft behavior from the UI interface to the underlying functional functions and effectively detect hidden malicious behaviors.