一种基于改进的贝叶斯投票算法的服务评估技术  

A Technique for Evaluating Services Based on Improved Bayesian Voting Algorithm

作　　者：张燕生[1] 白晓颖[2] 蒋长征[2] 

机构地区：[1]中国科学院研究生院,北京100049 [2]清华大学计算机系,北京100084

出　　处：《计算机科学》2008年第4期255-258,272,共5页Computer Science

基　　金：国家自然科学基金项目(编号:60603035)

摘　　要：现有的Web服务体系架构缺少服务质量与优选方面的有效支持,如何动态地选择、绑定并调用最适合用户需求的Web服务受到研究领域的关注。本文基于对"可信服务测试中介"模型的研究,提出了一种基于改进的贝叶斯投票算法的服务评估技术——把每一个测试评估结果作为对测试对象的一次投票,通过对协同服务测试数据的收集和分析,最终计算出同类服务的加权通过率,并在同类服务中进行排行,体现了Web2.0的技术特点。Web services architecture lacks of effective support for quality of service and Selection. It is an ongoing research direction in Web services community how to dynamically select, bind, and invoke Web service which can best meet the requirements of service consumer. Researching in the model of ＂Trustworthy Service Test Broker＂, it is proposed one technique for evaluating services based on improved bayesian voting algorithm on this paper. Every test result from collaborative test is thought as one pool for the test target. Through collecting, analyzing these collaborative test data, the pass rate of every service can be calculated, and then services of same type can be ranked. The technique has the characters of Web2. 0.

关 键 词：WEB服务 测试中介 协同测试 WEB2.0 服务排行 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP391.9[自动化与计算机技术—控制科学与工程]