Web service composition in the cloud environment Based on Modified Beetle Antennae Particle Swarm Optimization

摘要

Based on the fast development of Internet and big data, how to pick out a high-quality service composition quickly from large qualities of services, and meet users’ needs have become a hotspot in service composition research. Service composition optimization is a classic NP hard problem. A modified Beetle Antennae Particle Swarm Algorithm was proposed in this paper and it was applied in web service composition optimization. The author introduced Beetle Antennae Searching Algorithm into the modified Particle Swarm Algorithm, modified position update formula, adopted the modified Nonlinear Dynamic Trigonometric Learning Factors to control the particles’ expanding capacity and global convergence capability and incorporated the modified Secondary Oscillation factors which increased the searching precision of the algorithm and global searching ability. The experimental results under real data set indicated that the modified Beetle Antennae Particle Swarm Algorithm proposed in this paper could settle large-scale web service composition optimization problems in cloud environment, had good global searching ability, comparatively faster convergence speed and needed less time cost.