Description

With increasing presence and adoption of Web services on the World Wide Web, Quality-of-Service (QoS) is becoming important for describing nonfunctional characteristics of Web services. In this paper, the system present a collaborative filtering approach for predicting QoS values of Web services and making Web service recommendation by taking advantages of past usage experiences of service users. QoS value prediction of Web services is an important issue for service recommendation and selection. QoS value is an important factor for service selection because the quality of whole system is dependent on the QoS of single Web service. In reality predicting the missing QoS value of web service using the existing information is a difficult problem. Collaborative Filtering (CF) is one of the most widely used methods which employ QoS values contributed by similar users to make predictions. So the previous QoS value contributed by the different users is used for the prediction and the evaluation of missing QoS values. The reputation of the user is also an important factor for QoS value prediction. Existing Web service QoS value prediction approaches take data credibility into consideration, but did not take the location information, which may reduce the prediction accuracy. This paper presents two new approaches to solve the abovementioned problems of Web service recommendation. First the system propose a tensor-based QoS prediction method (TBQP), to predict multi-dimensional QoS accurately and easily. Then the system propose an overall QoS prediction method based on user preference learning (OQPUP), to obtain user preferences accurately and easily, thereby enabling us to accurately evaluate the overall QoS.