Active learning framework for android unknown malware detection

There are a lot of unknown labels Android examples in real-world applications, and they will cost much to mark manually. In this paper, we propose an active learning framework to solve this problem, so that Android malware are detected. In the active learning framework, Naive Bayes (NB), Decision Tree (DT), Logistic Regression (LR) and Support Vector Machines (SVM) are used to mark the labels for Android examples. The results indicate that this approach is effective to detect Android malware.