论文部分内容阅读
随着SOC(Service-Oriented Computing)的广泛应用,Internet环境下越来越多的资源以服务的方式进行封装.服务组合可以通过组合不同的资源来满足用户需求,而服务选择是服务组合的重要组成部分,其选择结果在很大程度上影响了组合服务的可信性.本文针对现有服务选择方法中存在的时间效率低下和查准率不高的问题,提出一种基于Petri网的Web服务选择方法,对服务库中的服务从功能和过程两个方面进行聚类预处理操作,缩小服务选择过程中的匹配范围,同时本文采用的基于Petri网的服务过程相似度计算方法,可以更准确地计算服务之间的相似度,进而在保证聚类效果的前提下提高服务选择的时间效率和查准率.仿真实验验证了本文所提方法的有效性.
With the widespread application of SOC (Service-Oriented Computing), more and more resources are encapsulated by service in the Internet environment.Service composition can meet user’s needs by combining different resources, and service selection is an important component of service composition Part of the results, the choice of results to a large extent affect the credibility of the combination of services.This paper aims to solve the problem of low efficiency and low accuracy in existing service selection methods, and proposes a Petri-Net-based Web Services Choose the method, carry on the cluster pretreatment to the service in the service storehouse from the function and the process two respects, narrow the match range in the service selection process, at the same time the Petri net based service process similarity calculation method that this text uses, can be more accurate To calculate the similarity between services, and then improve the time efficiency and accuracy of service selection under the premise of ensuring the clustering effect.The simulation results verify the effectiveness of the proposed method.