Research On Web Services Based On Functional Attributes And Non-functional Attributes

In recent years,with the widespread promotion and application of “Internet+”,Web service technology and service composition technology have been rapidly developed.More and more platforms provide Web services with the same or similar functions.How to meet user requirements for Web service calls better,faster,and more efficiently has become a research in academic circles and business circles at home and abroad.In most existing calculation models for Web service quality of service(Qo S),Qo S is usually analyzed using a constant-priority comprehensive evaluation mechanism,without considering whether the value of a certain value of the Qo S attribute is too large or too small,which leads to an overall evaluation of Qo S.Accurate,ignoring the interrelationships between Qo S attribute values may result in adverse effects on service selection.At the same time,because most of the current Web service composition problems are based on the local optimal principle,some old algorithms are used to solve the problem of service composition,such as mathematical programming method and exhaustive method,which greatly reduces the execution efficiency of service composition.Based on this,this article focuses on the hot and difficult issues in Web services and service composition,starting with the functional attributes and non-functional attributes of Web services,and analyzing the choice of Web services and service composition,mainly doing the following tasks and Innovation.First of all,we study the functional attributes of Web services,collect Web services on various service provision platforms,and sort and categorize Web services.By clustering methods,the services are divided into functional properties to generate Web services clusters with the same or similar functions.Then the requirements of the service demanders are matched with the Web service clusters to describe the matching results,and further follow-up to the Web services.The selection of non-functional attribute analysis and service combinations provides the premise and basis.Secondly,aiming at the deficiencies of the existing Web service Qo S attribute evaluation model,an improved method is proposed and a comprehensive evaluation mechanism of variable weight vector is introduced.Based on the comprehensive evaluation method of constant rights,the state variable weight vector is established to dynamically adjust the attribute weights of each index of service Qo S and improve the accuracy and objectivity of the evaluation of Web service Qo S attributes.Finally,aiming at the defects and deficiencies of traditional algorithms in Web service composition selection,a particle swarm optimization algorithm with linearly decreasing inertia weights and learning factors is introduced to better balance the self-cognition and social learning ability of particles and improve the particle's Search speed and global search capabilities.A large number of experiments were conductedand compared with traditional algorithms to verify the effectiveness and superiority of the improved particle swarm optimization algorithm.