MALDROID: A Hybrid Android Malware Detection System Using Machine Learning and Explainable AI

With the rapid growth of Android devices, the frequency and complexity of malware attacks have increased significantly. To address this challenge, we propose Maldroid, a hybrid detection system that integrates static permission analysis with dynamic behavior monitoring. Leveraging machine learning models such as Support Vector Machines (SVM), Neural Networks, and Ensemble Methods, Maldroid achieves precision rates exceeding 96%. Furthermore, by incorporating LIME (Local Interpretable Model-agnostic Explanations), the system provides interpretable results, thereby enhancing transparency and fostering trust in malware detection.