The Design And Implementation Of Travelling Service System Based On Reliability Prediction

With the development of Internet,more and more people are publishing their own services on the Internet in the form of Web Service,which are provided to users as open services.As a result,travelling service providers are gradually begining to stop developing their own travelling services and choose to use open Web services directly.At the same time,the demand of user for travelling service is becoming more and more comprehensive,and user often hope to obtain a one-stop service through an intermediary system.If these travelling services can be integrated into a travelling service system in the form of Web Service,it will not only improve service quality,but also save resources effectively.However with the development of Web services,the number and variety of travel related Web services are increasing rapidly,which results in the presence of Web services with similar functionality and different quality of service.Service quality will affect the user's experience of the system,and how to provide users with high quality service should be taken into consideration.Reliability is the most important and most obvious attribute of service quality,which can reflect the quality of travelling service system from the user level.As the services are connected through the network and run in a dynamic and changeable environment,the change of reliability will follow the change of time,and this kind of change does not have obvious regularity,therefore,how to predict the reliability of the service in order to optimize the service quality will be a needed problem to solve.This thesis studies the reliability prediction of service of travelling service system,and then designs and realizes the travelling service system based on reliability prediction.(1)This thesis proposed an optimization method of reliability prediction based on m_DBNs model,which can predict the reliability of travelling service in a dynamic and changeable environment.A time sequence collection generating method based on Lagrangian interpolation filling method is proposed to solve the problem of user's discontinuous use of service in time.A conditional probability table construction method based on k-means++ is proposed to improve its prediction accuracy and reduce its time cost.(2)This thesis designed and implemented the travelling service system based on reliability prediction,and applied the optimization method of reliability prediction based on m_DBNs model to the travelling service system.The system can be divided into four parts:service call module,service reliability prediction module,single category service recommendation module,service composition module.The response time and throughputcollected by service call module are used for the optimized reliability prediction method based on m_DBNs to get the predition result.The result will be the service selection basis of single category service recommendation module and service composition module.(3)This thesis verified the system.The thesis verified the function requirements of the system,and also the nonfuntion requirements,including reliability and efficiency,which are presented on the accuracy and efficiency of the optimization method of reliability prediction based on m_DBNs model.