Research And Design On Service Selection Approchers Based On QOS Attributes Filtering

Currently, there are massive Web services which QoS are high volatility on the Internet, which make users difficult obtain the high reliable Web services in a short period of time. Hence, how to fast and reliable carry out service selection based on Web service QoS attributes has become a research hotspot in the field of service computing.Although there are many existing service selection approaches could solve above problem, they are difficult to ensure the real-time and reliability of service selection due to ignore the redundancy of Web service and the uncertainty of their QoS attributes which lead to their service selection time consumption is too large and service composited results often deviate from the actual demand. Hence, this paper focuses on the real-time and reliability of Web service selection by reducing the search space of service selection and weakening the uncertainty of QoS attributes, and put forward the following three QoS-based Web service selection approaches according to different application scenarios.1. For the demand of users for high real-time of service selection, propose a fast Skyline Web service selection approach based on coefficient of variation. The approach first employs coefficient of variation to compute the QoS uncertainty of candidate services to filter higher QoS uncertainty candidate services. Then carry out Skyline computation and Skyline services solution for candidate services. Finally, fast select the best optimal Skyline composited services in which using0-1mixed integer programming meanwhile satisfy user’s global QoS constrains. 2. For the demand of users for high reliability of service selection, propose a reliable Web service selection approach based on variance. The approach first employs variance to compute the QoS uncertainty of candidate services to filter low reliable services and downsize the search spaces. Then compute the variance of QoS attributes of each candidate service according to a designed0-1mixed integer programming. Finally, select the high reliable composited services in which using a designed reliability function meanwhile satisfy user’s global QoS constrains.3. For the demand of users for high real-time and high reliability of service selection, propose a fast and reliable Web service selection approac. The approach first employs information and variance to compute the QoS uncertainty of candidate services to filter low reliable services and downsize the search spaces. Then design a reliability fitness function of service selection. Finally, fast select the high reliable composited services using0-1mixed integer programming meanwhile satisfy user’s global QoS constrains.