Credible Web Service Composition based on Improved Multi-objective Particle Swarm Optimization

Abstract

In the field of Web service composition, since the published value of Quality of Service (QoS) is not necessarily reliable, the research on the credibility of Web services has become a research hotspot. However, the current credibility research cannot well describe the difference between the published value and the actual operating value. At the same time, the algorithm used to solve the Web service composition scheme generally has the problem of poor solution effect. Based on this, this paper proposes a credible Web service composition method based on improved multi-objective particle swarm optimization (IMOPSO). Firstly, this paper proposes a similarity-based credibility calculation model, and combines it with the QoS published value through the inverse of the similarity, so that the QoS of the Web service meets the actual operation effect. Secondly, this paper proposes an improved multi-objective particle swarm optimization algorithm, which solves the problem of premature convergence by expanding the population diversity, and the solution effect is significantly improved. Finally, the improved multi-objective particle swarm algorithm is used to solve the Web service composition problem based on credible QoS. Simulation experiments on the QWS_V2 data set and the Synthetic data set prove that the method proposed in this paper can effectively improve the credibility of the service composition scheme, and has higher composition efficiency and better composition effect.