Web Service Composition Research Based On Artificial Bee Colony Algorithm

With the popularization of cloud computing technology,a large number of services with a single function have emerged in the network,but the needs of users usually require different services to work together.In order to meet the different needs of users,Web service composition technology emerges as the times require.Web service composition technology is to recombine candidate services with a single function in the resource pool to build composite Web services with more powerful functions and more application scenarios to meet user needs.On this basis,users further put forward requirements on the completion time,reliability and availability of the service.Therefore,how to choose a service combination that is satisfactory to users in the cloud computing environment is a hot topic of research nowadays,and it is also the research goal of this paper.This paper mainly focuses on how to use the artificial bee colony algorithm to solve the Web service composition optimization problem.The main research work is as follows:(1)Based on the quality of service(Qo S)requirements proposed by users: reliability,availability and service execution time,the weights are determined by the AHP,and a multi-attribute combined service evaluation model is constructed according to the basic topology structure of Web services.Transform the Web service composition optimization problem into a single-objective optimization problem.Due to the differences between services in the dataset,in order to improve the efficiency of service combination and selection,this paper introduces and improves the PROMETHEE II algorithm for the preprocessing of the dataset.(2)Aiming at the shortcomings of artificial bee colony algorithm(ABC),an enhanced artificial bee colony algorithm(EABC)is proposed: 1)The adaptive method based on opposition learning is added to the ABC,which increases the interference in the early stage to prevent the algorithm from converging prematurely,the interference will gradually disappear in the later period without affecting the final convergence of the algorithm;2)Introduce a chaotic strategy in the onlooker bee stage and add the current global optimal solution.The chaotic strategy can improve the random search ability of the onlooker bees,and improve the local search ability near the high-quality solution according to the nectar source information provided by the employed bees.At the same time,adding the global optimal solution and exploring its nearby search space can improve the development performance and accelerate the convergence of the algorithm;3)Improve the assignment strategy of the scout bee stage,assign a high-quality solution to the stagnant bee and add interference.(3)The optimization method proposed in this paper is simulated.Differential evolution algorithm,particle swarm algorithm,empire competition algorithm and ABC were set as control groups,and the preprocessed data set and the original data set were used to solve the user evaluation of Web services of different scales.From the effectiveness,convergence,stability and execution time of the algorithm,the performance difference between the comparison algorithm and the EABC algorithm is compared.The simulation results show that the data set preprocessing is effective,and the improved artificial bee colony algorithm has stronger comprehensive ability.After combining the two,the execution time of the EABC algorithm is significantly reduced,and the obtained fitness value is better and more stable.