服务搜索引擎中基于接口匹配的服务发现功能的设计与实现

发布时间：2018-04-26 14:05

本文选题：Web服务 + 服务发现　；参考：《北京邮电大学》2013年硕士论文

【摘要】：互联网经过多年的发展,已经日渐普及。一提到网络,人们就会想到Web, Web确实为互联网的发展做出了巨大的贡献。Web服务因其对异构平台应用程序互操作提供支持,成为开发Web应用的重要平台。伴随Web服务的井喷式发展,互联网上可用的Web服务越来越多。如何使用户从大规模Web服务注册库中快速、准确地查找所需要的服务,是Web服务继续发展必须要解决的问题,也成为了国内外研究的热点。起初,Web服务发现机制基于UDDI (Universal Description, Discovery and Integration)。UDDI是一种目录服务,支持基于关键字的查找方法,仅停留在语法层面,缺乏语义推理的支持。为此,研究者纷纷引入语义Web的思想,试图提高Web服务发现的查全率和查准率。由于Web服务的语义化需要更改服务实现的支撑协议,而且很多技术尚不成熟,所以不能用于商业运行。在基于关键字的服务发现机制和基于语义的服务发现机制之外,本文依托课题“融合网络业务体系的开发”项目第一阶段提出了一种新的服务发现机制。该机制把Web服务抽象为具有输入输出参数的接口实体,通过匹配接口信息查找相符的Web服务,称之为基于接口匹配的服务发现机制。在多维度的接口信息中,综合多种决策依据进行服务匹配,需要计算多种匹配形式。项目第一阶段已经对接口的各个描述信息的匹配算法进行了研究,但其研究工作有待于进一步完善。本文总结归纳出匹配形式可以有参数个数匹配、参数次序匹配、参数类型匹配、参数名称的关键字匹配和语义匹配。因此,对每一种匹配形式的设计是服务接口匹配的关键技术。在参数类型匹配模型设计中,考虑到Web服务支持跨平台的数据类型系统,接口的参数其实具有很多类型,它可以是内置数据类型,也可以是自定义的类型,可以是简单数据类型,也可以是复杂数据类型。对于复杂类型的匹配,本文引入了图论中的二分图模型。首先对其成员数据类型进行原子化操作,然后通过二分图匹配的方法判断其与目标复杂参数类型的匹配度,这是本文的重要研究内容。此外,基于接口匹配的服务发现功能的设计与实现也是本文主要讨论的内容。作为服务搜索引擎的一部分,服务接口匹配子系统实现了该服务发现功能。本文在明确系统功能需求的基础上,首先对其进行总体设计,然后划分子系统的功能模块,接着设计了各模块的操作流程,最后通过测试用例对其进行验证。最后,本文对已经取得的研究成果进行总结,找出存在的问题,探讨了可供继续研究的问题,指出了下一步的研究工作。
[Abstract]:After years of development, the Internet has become increasingly popular. When it comes to the network, people will think that Web has made a great contribution to the development of the Internet. Web services have become an important platform for developing Web applications because of its support for the interoperability of heterogeneous platform applications. With the blowout development of Web services, more and more Web services are available on the Internet. How to make users find the required services quickly and accurately from large scale Web service registries is a problem that must be solved for the continued development of Web services and has become a hot research topic at home and abroad. At first, the mechanism of web service discovery is based on UDDI Universal description. Discovery and Integration).UDDI is a kind of directory service, which supports keyword-based lookup method. It only stays at the level of syntax and lacks the support of semantic reasoning. Therefore, researchers have introduced the idea of semantic Web to improve the recall and precision of Web services discovery. Because the semantics of Web service needs to change the supporting protocol of service implementation, and many technologies are not mature, it can not be used in commercial operation. In addition to the keyword-based service discovery mechanism and the semantic-based service discovery mechanism, this paper proposes a new service discovery mechanism in the first phase of the project "integrating the development of network service architecture". The mechanism abstracts the Web service as an interface entity with input and output parameters, and finds the matching Web service by matching the interface information, which is called the service discovery mechanism based on interface matching. In the multi-dimensional interface information, it is necessary to calculate a variety of matching forms by synthesizing various decision bases for service matching. In the first phase of the project, the matching algorithms of each description information of the interface have been studied, but the research work needs to be further improved. In this paper, we conclude that the matching forms can include parameter number matching, parameter order matching, parameter type matching, keyword matching of parameter name and semantic matching. Therefore, the design of each matching form is the key technology of service interface matching. In the design of the parameter type matching model, considering that Web services support a cross-platform data type system, the parameters of the interface actually have many types, which can be either built-in data types or custom types. It can be a simple data type or a complex data type. For the matching of complex types, this paper introduces the bipartite graph model in graph theory. Firstly, the atomization of its member data types is performed, and then the matching degree between them and the complex parameter types of the target is judged by the bipartite graph matching method, which is an important research content in this paper. In addition, the design and implementation of service discovery based on interface matching are also discussed in this paper. As a part of the service search engine, the service interface matching subsystem implements the service discovery function. On the basis of defining the functional requirements of the system, this paper first designs the overall design of the system, then divides the functional modules of the subsystem, then designs the operation flow of each module, and finally verifies it by a test case. Finally, this paper summarizes the research results, finds out the existing problems, discusses the problems that can be further studied, and points out the next research work.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP393.09

【参考文献】