Optimal Malware Detection for Android

Abstract—The increasing prevalence of Android devices has led to a surge in malicious apps targeting the platform. We present an Automated Android Malware Detection system using an Optimal Ensemble Learning Approach to combat this. This system integrates machine learning algorithms like Random Forest, Gradient Boosting, and Convolutional Neural Networks to improve detection accuracy and efficiency. Our ensemble learning technique combines predictions from multiple models, and we introduce a novel feature selection method tailored for Android malware detection. Tested on a large dataset of malware and benign apps, our approach outperforms existing methods in accuracy, false positive rate, and computational efficiency. This system effectively handles the evolving nature of Android malware, providing a practical solution for enhancing device security and protecting users from threats. Keywords—Ease of user experience, Enhanced customer service, Voice recognition, user feedback, Time efficiency.