Research On The Particle-Ant Colony Algorithm Based On QoS In Web Services Composition Problem

As the internet technology developing and the quantity of the applicationsincreasing, there are more and more shared Web services existed in internet.However, these Web services are fine-grained simple services, can not meet theusers’ demand. So we need one technology to achieve services being reused andsaving the resource of the internet.In order to improve the efficiency of finding the optimal solution for Webservices composition problem to meet increasingly sophisticated demand, analgorithm called Particle-Ant Colony Algorithm (PACA) based on quality of service(QoS) is proposed. In this article, we introduce the technology, the algorithm and themodel in details which classifys atom-Web-services that shared in internet into Webservice candidate sets and solves Web services composition problem.Firstly, classify these Web services into several Web service candidate sets(WSCS) according to the similarity of two services’ name and the input-outputparameters in WSDL file. Then, build abstract model of Web service compositionproblem.Secondly, converts the Web services composition problem into QoS-baseddirected acyclic graph shortest path problem.Thirdly, propose Particle-Ant Colony Algorithm (PACA) after research theParticle Swarm Optimization Algorithm (PSOA) and Ant Colony OptimizationAlgorithm (ACOA). The main idea of the PACA is: find several suboptimal paths byPSOA and initialize the pheromones of these paths firstly, then find optimal solutionby ACOA.Finally, apply PACA into Web service composition problem and built simplemodel of Web service composition problem, and compare the experiment result thatPSOA and ACOA get under the same experiment environment.The experiment resultindicates that the PACA can effectively improve the capability of finding optimal solution of Web services composition problem.