The Research Of Web Service Selection Based On QoS And Uncertainty User's Preference

With the continuous improvement of the relevant standards of Web service, enterprise application software platform technology is becoming more mature, Web service in many areas has been widely used. Enterprise in order to expand their operational capacity and enhance the core competitiveness typically the core business and the business processes encapsulated into standard Web service and registered and published on the UDDI. How to call from the UDDI service database select service and bind service technology become a hot topic of Web service. Since many Web service in network, QoS-based service selection method to select Web service has become a hot topic. This paper studies the QoS-based Web service selection technology, which is currently the main research service selection technology. The concerned QoS attribute in Web service selection method main focus is the service QoS attribute. In service selection, or ignore or assum the role of the user in the service selection process usually. For these reasons, the user's wanted service compared to the select service in service selection method with gread difference. Therefore this paper from the user's point and give full consideration to the role of service users in the service selection proposed the service selection method based on QoS and uncertainty user's preferences.The main contents of this paper as follows:(1) In the full study Web service technology, introduces the background and significance of Web service technology, then the status of service selection technology research at home and abroad, introduces the associate concept and implementation technique. Finally, give the calculate QoS model.(2) As the disadvantage in the result service of existing Web service selection to select service compared to the user's wanted service has great difference. Therefore, this paper on the basis of gray correlation theory and fuzzy set theory proposed based on QoS and the uncertainty user's preference Web service selection method to solve the above drawbacks. So that the result of the service selection to maximize approximate user wanted. For shortcomings of the single weighted strategy and poor scalability in traditional Web service selection, this paper give a weighted strategy based on the correlation number, to ensure the weight assignment stretchable, so that the weight calculation more comply with the service selection in actual situation.(3) The existing Web service selection method of single function and does not consider the missing of QoS attributes information, thus inconvenient the user to select service. Therefore, to solve the missing of QoS attribute information in service selection, we propose the missing QoS attribute information prediction method based on grey correlation theory. By this method combined Web service selection method can further improve the accuracy with service for service selection. Through experiment we found that the proposed prediction method compared to other methods has higher prediction accuracy within a certain range.