Efficient Mining of User Behaviors by Temporal Mobile Access Patterns

Ubiquitous computing offers various kinds of dynamic services to the mobile users with versatile devices at anytime and anywhere. In ubiquitous environment intelligent mobile agents are mandated to communicate with users and it is enabled by capturing interesting user's behavior patterns. Existing mining methods have proposed frequent mobile user's behavior patterns statistically based on requested services and location information. In this case some problems are caused because it was not considered that the mobile user's dynamic behavior patterns are usually associated with temporal access patterns. Therefore, ubiquitous computing provides the dynamic services with timely manner when user wants to get useful services information on time. In this paper, we propose a novel data mining method, namely temporal mobile access patterns that can efficiently discover mobile user's temporal behavior patterns associated with location and requested services. Furthermore, we present a novel data structure T-Map to store the temporal mobile access patterns. The advantage of our data structure compactly stores the user's behavior pattern according to location and service information in memory. Even though the information data sets require large shared memory when they are stored, our approach still provides fast access and consume less memory than other methods. The proposed technique in this paper works especially well for context-awareness data sets in mobile agent systems.