基于不确定性QoS的Web服务选择方法研究

发布时间：2018-05-14 05:39

本文选题：Web服务 + QoS(服务质量)　；参考：《重庆大学》2014年硕士论文

【摘要】：Web服务因具有跨平台性、松散耦合性、互操作性、可扩展性等优点受到国内外专家和学者的共同关注。随着网络上部署的Web服务数量的增加，出现了许多功能相同的Web服务，当人们使用Web服务的时候很难确定哪些服务最适合自己的需求。因此，Web服务选择成为了Web服务应用过程中的关键环节。本文主要研究基于QoS的Web服务选择。目前基于QoS的Web服务选择都只考虑了给用户提供服务选择的结果而没有给服务提供者提供改进服务的反馈信息，本文提出了面向服务提供者的服务质量反馈机制；对QoS属性计算时，现有的方法中QoS属性取值也较为单一，仅采用实数类型表示，QoS属性值不支持区间型和语言型表达。区间型数据能描述数据的波动情况，反映数据的稳定性特征，语言型符合人们日常表达的习惯。而对用区间数和语言型等不确定性数据表示的QoS模型进行服务选择成为了Web服务选择领域的难题之一，本文提出了基于不确定性QoS的Web服务选择方法。本文的主要工作：（1）本文首先介绍了Web服务选择的研究背景与意义并对服务选择技术的国内外研究现状进行了分析，接着介绍了Web服务相关的基础知识，并给出了Web服务选择的具有不确定性QoS的计算模型。（2）现有的Web服务选择都只提供了服务选择的结果，，而没有给服务提供者提供反馈。本文提出了面向服务提供者的服务质量反馈机制，使用户能在选择到自己满意的服务的同时也让服务提供者定性的掌握他们所提供的服务非功能属性的好坏情况，为服务提供者提供了改进服务的依据。（3）Web服务的非功能属性QoS在大部分的研究中采用的都是数值类型，其类型结构比较单一。本文提出了基于不确定性QoS属性值的Web服务选择方法（即QoS属性值表示成实数、区间数和语言型等形式），增强了QoS模型的描述能力。本文将基于区间数QoS的Web服务选择采用基于区间数可能度的方法实现，完全的保留了区间数的特征；对基于混合QoS的Web服务选择中的语言型表示的QoS转换成三角模糊数保留其不确定的特性。最后以WS-DREAM实验数据为基础，通过实例和仿真实验验证了方法的有效性和正确性。（4）本文还使用了基于多个用户访问Web服务的历史数据进行服务选择的方法，分别使用基于区间型QoS模型和基于混合QoS模型进行服务选择。综合考虑多个用户使用服务的情况再做出服务选择，使服务选择的结果更加可靠。
[Abstract]:Because of its advantages of cross-platform, loose coupling, interoperability and scalability, Web services have attracted the common attention of experts and scholars at home and abroad. With the increase in the number of Web services deployed on the network, there are many Web services with the same functions. When people use Web services, it is difficult to determine which services are most suitable for their own needs. Therefore, the selection of web services has become a key link in the application of Web services. This paper focuses on Web service selection based on QoS. At present, Web service selection based on QoS only considers the result of service selection to users, but does not provide the feedback information of improving service to service provider. In this paper, a service provider oriented service quality feedback mechanism is proposed. When calculating the QoS attribute, the value of the QoS attribute in the existing methods is also relatively simple, and only the real number type is used to indicate that the value of the QoS attribute does not support interval and language expression. Interval data can describe the fluctuation of data and reflect the stability of data. However, the selection of QoS services based on uncertain data such as interval number and language type has become one of the difficult problems in the field of Web service selection. In this paper, a method of Web service selection based on uncertain QoS is proposed. The main work of this paper is as follows: Firstly, this paper introduces the research background and significance of Web service selection, analyzes the current research situation of Web service selection technology at home and abroad, and then introduces the basic knowledge related to Web services. An uncertain QoS computing model for Web service selection is also presented. The existing Web service choices only provide the results of the service selection, but do not provide feedback to the service provider. In this paper, a service quality feedback mechanism for service providers is proposed, which enables users to select satisfactory services and to qualitatively master the non-functional attributes of the services they provide. It provides a basis for service providers to improve their services. The non-functional attribute QoS of web services is a numerical type in most researches, and its type structure is relatively simple. In this paper, a Web service selection method based on uncertain QoS attribute values is proposed (i.e., QoS attribute values are expressed as real numbers, interval numbers and language types), which enhances the description ability of QoS model. In this paper, the Web service selection based on interval number QoS is implemented based on the possibility degree of interval number, and the characteristic of interval number is completely preserved. The QoS transformation from linguistic representation in Web service selection based on mixed QoS to triangular fuzzy number retains its uncertainty. Finally, based on the experimental data of WS-DREAM, the validity and correctness of the method are verified by examples and simulation experiments. 4) this paper also uses the method of service selection based on the historical data of multiple users accessing Web services, using interval-based QoS model and hybrid QoS model for service selection, respectively. Considering the service usage of multiple users, we can make the service selection more reliable.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.09

【参考文献】