Machine Learning Models for Android Malware Detection–A Comparative Study

In the mobile ecosystem, Android apps are expanding quickly, but malicious software for Android is also proliferating in ever-increasing quantities. The issue of identifying malware on Android has been examined by several researchers, who have proposed ideas and approaches from various angles. According to current research, machine learning (ML) is a promising and successful method of detecting Android mal ware. This study is performed to compare the various ML methods for Android malware detection, including Random Forest (RF), Decision Tree,(DT) Random Tree(RT), Support Vector Machine (SVM), and XGBoost. The study assessed these models using a Androzoo dataset of Android apps, taking into account the important performance parameters as detection latency, accuracy, precision, and false alarm rate (FAR).