Optimization of Accuracy Using Ensamble Stacking to Detect Android Malware

The rapid growth of Android devices has led to an increase in malware targeting this platform, necessitating advanced detection methods to protect users. This study focuses on developing an ensemble stacking model for Android malware detection, utilizing the Drebin-215 dataset. The proposed model integrates multiple base classifiers-Support Vector Machines, Random Forest, Logistic Regression, K-Nearest Neighbors, AdaBoost, and Gradient Boosting Machine-to enhance detection accuracy. Logistic Regression is employed as the meta-classifier to effectively combine predictions from these diverse classifiers. The performance of the model is evaluated based on accuracy, precision, recall, and F1-score, demonstrating its robustness in detecting various types of Android malware. The findings suggest that the ensemble stacking approach, even without advanced feature selection and optimization techniques, can significantly improve the accuracy and reliability of Android malware detection systems.