A User-Centric uncertain data model for homogeneous service

In the web environment, there are more and more web service resource available, and in each application domain, like news, travel, weather, and so on, there are a number of web services sharing the same functionality which we call them homogeneous service. As the homogeneous services give us more choices, we also face the uncertainty problem that these homogeneous services always provide inconsistent data results for the same entity because of different information sources and this kind of data uncertainty can´t be solved by a single service. So we need to describe the uncertainty of homogeneous services. In this paper, we propose a User-Centric uncertain data model for Homogeneous Service. We use probabilistic model to describe the uncertainty of service, assess the uncertainty based on the QoS metrics of service which describe the objective reliability and user´s decision of service selection which reflect the subjective reliability. Through the experiment, the results show that our method can effectively combine the information of subjective reliability and objective reliability to reflect the data uncertainty of the service.