An Effective Transactional Service Selection Approach with Global Optimization Based on Skyline and Particle Swarm Optimization

Abstract

With the growing number of alternative services in the open service environment, service selection with global optimization in service composition is a critical issue. In this paper, we propose an approach SPSO-GOTSS (global optimization of transactional service selection based on skyline and particle swarm optimization) to implement transactional service selection with global optimal QoS and semantic matching degree. This approach first adopts skyline operator to prune redundant services, then employs particle swarm optimization to select service from amount of candidates. When computing the final skyline service, we consider both dominance and incompatibility checking. The mutation operation is used to overcome the premature convergence of traditional PSO. The experimental results show that our proposed approach is feasible and effective.