Automated Android Malware Detection Using Optimal Ensemble Learning Approach for Cybersecurity

Abstract

The proliferation of Android malware poses significant threats to mobile security, necessitating advanced detection mechanisms. Traditional static and dynamic analysis methods often fall short due to the evolving nature of malware and the vast diversity of applications. This paper proposes an automated Android malware detection framework employing an optimal ensemble learning approach to enhance detection accuracy and robustness. The framework integrates multiple base classifiers, each trained on distinct feature sets, to capture a comprehensive range of behavioral and structural patterns inherent in Android applications. A meta-classifier is utilized to synthesize the outputs of base learners, optimizing the final decision-making process. Experimental evaluations demonstrate that the proposed system outperforms traditional single-model approaches, achieving higher detection rates and lower false positive rates across various datasets