Abstract-The diverse types of mobile applications are used regardless of time and place, as a number of Android mobile device users have been recently increased. However, the breach of privacy through illegal leakage of personal information and financial information inside mobile devices has occurred without users' notices, as the malicious mobile application is relatively increasing In order to reduce the damage caused by the malicious Android applications, the efficient detection mechanism should be developed to determine normal and malicious apps correctly. In this paper, we aggregated real-time system call events activated from malware samples distributed by Android Malware Genome Project. After extracting the basic difference feature and characteristics of system call events pattern from each normal and malicious applications, we can determine whether any given anonymous mobile application is malicious or normal one.