AndroPack: A Hybrid Method To Detect Packed Android Malware With Ensemble Learning

In the rapidly evolving landscape of mobile security, the increasing risk posed by Android malware is a paramount concern. The inherent openness of the Android system emphasizes the critical need for continuous vigilance and strategic alertness. Packed malware poses a significant challenge in detection and analysis, as these deliberate techniques complicate traditional security measures, allowing malicious actors to conceal their intent effectively. This research proposes an automated system that integrates static and dynamic analysis, leveraging sophisticated methods to extract pertinent features from the APK. Additionally, this research incorporates an ensemble learning framework to enhance the robustness and accuracy of the analysis. The proposed system comprehensively examines the complexities of packed Android malware, thereby increasing detection accuracy by 95% and reinforcing the security infrastructure in mobile environments, addressing the dynamic nature of mobile threats.