With the sharp increase in the number, smartphones are now ubiquitous. However, the security requirements of these new systems and the applications which they support are still being understood. Android platform, as a market leader, makes the need for malware analyses on the platform become an urgent issue. In this paper, a data mining approach through dynamic analysis of application behavior for detecting malware in the Android platform is proposed. We established a framework for collection of traces from an unlimited number of real users based on cloud computing, and behavior-related dataset of each application will be created. Then the partitional clustering algorithm is used to cluster each dataset. Through experimental results, the approach is shown to be an effective means of detecting the malware.
