An efficient approach to web service selection

Abstract

Under the service-oriented circumstances, developers can rapidly create applications through Web service composition. Owing to existing service selection methods having close relationship with the available service number, the efficiency inevitably decreases with the increase of available services. In the paper, a discrete Particle Swarm Optimization (DPSO) oriented to efficient service selection is designed according to Particle Swarm Optimization (PSO) idea. Furthermore, aiming at the practice service composition requirements, an algorithm used to resolve the dynamic service selection with multi-objective and quality of service (QoS) global optimal is presented based on DPSO and the theory of intelligent optimization of multi-objective PSO. Theoretical analysis and experimental results show that DPSO and multi-objective DPSO (MODPSO) are both feasible and efficient in the process of service selection. © 2009 Springer-Verlag.