摘要：New challenges about data services have arisen, especially consider the impact of user behavior. For dealing with the problem of distributed heterogeneous data sharing and satisfying the demands of data service, the complex association and dynamic changes should be tracked timely. Thereby it has become increasingly important to build a data services framework based on the user behavior analysis. In this paper, we propose an effective user behavior modeling approach based on the Open Cloud Service Architecture (OCSA) to manage the domain scientific data. Firstly, we build the data services framework in the cloud environment, and describe the related theories and conceptual model about this framework. Based on this, we elaborate the definition and classification of user behavior. Secondly, we construct the User Behavior Model (UBM) by tracking the user operation behavior respectively from the time dimension and the space dimension. Finally, we designed and realized a behavior-oriented Materials Scientific Data Sharing Service Platform. Through the description and comparison of application case, the validity of user behavior modeling method is verified.
