【作者】 安杨；

【导师】 边馥苓；

【作者基本信息】 武汉大学 ， 摄影测量与遥感， 2005， 博士

【摘要】 Web服务模式为分布式地理信息系统(GIS)提供了新的机遇,也提出了新的挑战。然而,现有的Web服务的行业标准以及GIS领域中的相关规范由于不能很好的表达Web服务的语义信息和解决Web服务应用中的语义冲突,难以满足Internet环境下网络地理服务应用的需要,因而寻求新的网络地理服务模式是高效、实时的应用网络地理服务的一个重要研究方向和前沿课题。 本体具有良好的概念层次结构和对逻辑推理的支持。本体作为共享概念模型的形式化规范说明,能够准确的捕获相关领域的知识,提供对该领域知识的共同理解,确定该领域内共同认可的词汇,并从不同层次的形式化模式上给出这些词汇之间关系的明确定义。基于这种考虑,本文将本体和Web服务技术相结合,提出了建立基于本体的网络地理服务模式的思路,并对其中若干关键问题,即基于本体的服务描述、发现和合成等相关的理论和技术问题进行了广泛和深入的研究。 本文提出了一种基于本体的网络地理服务模型和基于该模型的系统体系结构,研究了模型中的网络地理服务描述、发现和合成等关键理论和技术问题,并取得了如下的研究成果: 1.在对当前Web服务的行业标准及网络地理服务的规范进行详细分析和对本体进行了深入研究的基础上,提出了一种基于本体的网络地理服务模式的新思路。该模式主要由五个部分构成:网络层、网络地理服务发布层、网络地理服务发现层、网络地理服务合成层和应用层。网络层位于最底层,包含了标准网络协议(如:HTTP、SMTP和IIOP),允许通过标准方法进行对象描述交换的SOAP协议,以及支持机器Agent进行通信的协议。网络地理服务发布层是用于网络地理服务发布,发布的服务是以服务描述的形式被存放到发布层的服务注册中心。网络发现层是由若干个匹配器构成,用户可以通过各种方式在网络地理服务发现层中查询和调用该服务。网络地理服务合成层将Internet上已有的相对简单的地理服务按照一定的业务流程逻辑合成起来以构成复杂的复合服务完成用户所提出的复杂的任务。应用层包括需要获取网络地理服务的应用程序和人类用户,它还提供不同的GUI界面以便用户提交查询要求和进行服务结果的显示。在整个模式中,网络地理服务的描述、发现和合成均是基于本体的。本体的引入给网络地理服务带来新的特性,使其能够更好的适应开放、异构的Internet环境,以实现分布式Internet中各种地理服务的互操作。 2.针对目前的服务描述方法并不能很好地表达Web服务的语义信息,所描述的Web服务是静态的,不包含任何有关服务执行过程的信息,本文提出了基于本体的网更多还原

【Abstract】 Web Services as a promising research field brings new opportunities and challenges to distributed GIS. However, industry standards of Web Services and OGC specification make Web Services incompetent to fulfill various requirements of the coming GIS applications under dynamic environment. The main inhibitor is the lack of semantics in the description of Web Services and discovery process. Therefore, new model of Geo Web Services is an active area of research in GIS.As a recently developed paradigm in semantic web, Ontology can especially benefit Geo Web Services applications on providing the semantics of services and requestor’s requirement. Ontology is an explicit specification of a sharing conceptualization, which can be used in an integration task to describe the semantics of information sources and to make the contents explicit, also can be used for the identification and association of semantically corresponding information concepts. Considering the advantages of Ontology and Web Services technology, we propose a new paradigm in constructing Geo Web Services with the combination of both technologies, and further explore the related issues of its principles and techniques.This thesis presents the Geo Web Services model based on Ontology and architecture based on model, and addresses the fundamental principles and implementing techniques of Ontology in the model, describes a prototype. Major contributions of the thesis include: 1. A new paradigm of Geo Web Services based on Ontology is proposed based on analyse of industry standards of Web Services and OGC specifications. The introduction of Ontology to Geo Web Services brings new merits to make it more adaptable to open and dynamic Internet environment. The Geo Web Services model based on Ontology includes five layers: Network layer, Geo Web Services Publication Layer, Geo Web Services Discovery Layer, Geo Web Services Composition Layer and Application Layer. The Network Layer forms the lowest layer in the architecture and encapsulates standard network protocol (such as HTTP, SMTP and HOP) and SOAP allowing for exchange of object descriptions by standard means. In addition, machine agents need the possibility to communicate at an adequate level using protocols such as Knowledge Query and Manipulation Language (KQML) or Agent Communication Language (ACL).The Geo Web Services providers create Geo Web Services and their services definition that are based on Ontology, and then publish the services with the Geo Web Services registry based on OWL-S (an Ontology of service). Once a Geo Web Service is published, a service requestor may find the services via the registry. Geo WebServices Discovery Layer is used to find the appropriate Geo Web Services for the requestor’s task. Geo Web Services are advertised using Ontology, which is used in service matchmaking too. Geo Web Services composition can be interpreted as pipelining services in a dependent series to attain a customized, more complex request. And this layer is responsible for carrying out the process of managing the discovery and integration of services to construct a composite service. The application layer embodies any application and human user that request the Geo Web Services. The application layer encompasses different GUI facilities to display the result of Geo Web Services and provides the functionality to initiate a request for Geo Web Services.2. Description architecture of Geo Web Services is proposed, which contains two levels: application level and conceptual level. Application level is Geo Web Services description based on OWL-S, which is given by Geo Web Service provider. The elements used in OWL-S are annotated with concepts from Conceptual level. The Conceptual Level is the level of human concepts and provides an agree-upon conceptualization by GIS Domain Ontology. The domain ontology makes a world view explicit and presents it in a machine interpretable way. Such domain ontology will never be complete since the meteorological knowledge evolves and the terms used change. But it will serve to describe a particular worldview on this domain and the vocabulary humans use to communicate about it. And domain expert should participate in the building of domain Ontology. This description architecture provides the Geo Web Services description of functional, non-functional and behavioral information in a semantic as well as in a syntactic form of representation. It also supports inferences on descriptions. This means that automated reasoning and comparison on descriptions will be possible and efficient.3. Geo Web Services discovery based on Ontology is proposed. The matchmaker is implemented based on matchmaking algorithm. The service provider advertises services to Registry. When matchmaker receives the requirement description of requestor, the services advertisements in Registry will be sent to matchmaker for capability matching. The result of the matching is the list of advertisements that are satisfied with requirements. Ontologies database contains ontologies of general knowledge, domain ontologies and knowledge concept ontologies, which are used to judge relation of elements of two descriptions. Matchmaker adopts techniques of information retrieve, AI and software engineering to calculate similarity degree between Geo Web Services description and requirements. This method can solute semantic conflict in traditional discovery and improve efficiency of discovery.4. A new dynamic service composition based on Ontology is proposed. Each workflow thatstored in the workflow repository is composed of one or more web services, which can be situated anywhere on the web. The OWL-S descriptions of the Geo Web Services are queried from the Registry, and the service matching that described above is applied. The workflow composer agent composes the executable workflow that comes from workflow repository based on composite process, and sends it into the workflow execution engine. Finally, the execution engine executes the workflow using the Geo Web Service instances.5. A simplified prototype of Geo Web Services based on Ontology is developed. Currently the system is implemented by using Protege3000 as platform for Ontology. User can describe, discover and compose Geo web services through this system. The success of the prototype shows that it is feasible to establish distributed GIS with Geo Web Services based on Ontology.更多还原