Malware Detection in Android Devices Using Histogram-based Gradient Boosting

This research is to develop a method of using machine learning to make Android malware detection more accurate. Eleven thousand five hundred eighty-eight samples of Android that are of five types of malware were used. There was a lot of work that had to be done in the data preparation that included using tools such as SMOTE to balance and select features. The classifier that is histogram-based gradient boosting (HGB) was the most effective algorithm following the many tests of machine learning algorithms. The HGB model's accuracy is 99.8 percent, which is unbelievably high compared to other models like Decision Tree, XGBoost, and Random Forest. The findings of this study emphasize the necessity for the use of complex machine learning methods and proper data processing to increase the malware detection precision. The suggested method is not only a security method that allows Android malware to be detected but also a new standard for the research in this area.