A Machine Learning Approach for Detection of Malicious Applications in Android

Abstract

In this paper, a machine learning approach for detection of malicious applications in an Android operating system is proposed. In today’s world, android is a largest operating system used by peoples around the world and its users are still growing day by day. As android is used on a large scale by its users and due to its popularity, the malicious applications are also growing to harm the system or to use the personal information of an android user. A recent report by Google said that the company detected 99 percent of apps with malicious content before anyone could install them. The implementation of malicious app detection tool analyze the relationship between system functions, sensitive permissions and sensitive application programming interfaces. The different machine learning technique, Grid Search and SVM Algorithm is performed to analyze the result that can detect the rate of malicious applications with a better efficiency and can improve the performance of the system with the machine learning methodology.