【作者】 王越；

【导师】 万建成；

【作者基本信息】 山东大学 ， 计算机技术， 2007， 硕士

【摘要】 当前互联网的规模越来越大，但网络资源缺乏统一的语义描述，因而计算机难以有效理解Web上信息的含义，用户越来越难以从海量的数据中让计算机查找和处理自己需要的资源，同样的问题也存在于政务Web信息资源处理领域。所以笔者首先在本文中具体分析了当前电子政务信息资源处理中存在的这一问题。如政务信息资源共享程度低、互操作性差、关键词检索效果差、信息处理缺乏智能化和人性化的支持问题等等，而W3C组织倡导的基于本体的语义网技术为解决这些问题提供了出路，语义网通过获得词汇的语义，可以使政务Web应用不但能够表示信息，而且能够理解信息，还可在此基础上依据一定的规则对信息进行推理。随后本文介绍了语义Web的相关理论和技术概念，包括语义Web的概念、研究现状、体系结构和关键技术，本体的概念、类型、描述语言、构建方法和编辑工具等等。在分析了语义web和本体论的基础技术之后，本文提出了一个政务信息资源本体的获取模型，该模型以本体构建的循环获取方法论为理论指导，借助《电子政务主题词表》转化生成电子政务信息资源原始领域本体，并且在领域专家或系统管理者审核的前提下，领域本体可以在以后的语义信息提取过程中持续学习新概念实例，不断完善，从而使得以上两种本体构造方法能以一种互补互利的方式结合运用。随后根据中文自然语言信息处理的特点，并参照上述互补的本体构造方法，本文提出了一个Web政务资源的语义信息提取和标注模型，在对模型所用到的概念和参考资源作了分析之后，本文对模型中的中文语言的语义分析提取和网页语义标注这两个重点问题做了详细讨论。最后，本论文基于上述技术分析，对两个基于Ontology的Web政务资源的智能处理问题进行了初步的模型探讨，即市长信箱分类处理系统架构和语义政府门户网站的架构。更多还原

【Abstract】 Nowadays web is growing amazingly, but the information lacks a uniform semantic description, so it is difficult for computer to understand information on web, and it is increasingly difficult to find, organize, access and maintain the information required by users. The same thing happens in the e-government information resources processing areas.So first the author analyzed the problems of the processing of government information Resources at present, such as low degree of sharing, bad effect of keyword search and lack of intellectualized and humanized support. Ontology-based Semantic Web technology proposed by W3C indicates a way to solve this problem. E-government web applications can represent and understand the information by obtaining the semantic of the words, even though infer from the information based on some roles.And then the paper introduces the related theory and technology of Semantic Web. including the concept、 research status、 architecture and key technology of Semantic Web, the concept、 types、 description languages, building and edit tools of Ontology.The paper studies the relevant problems in the process of ontology acquisition, including transforming e-government subject key word into domain ontology, and under the audit of e-government domain expert or ontology administrator, e-government domain ontology can continuously learning new ontology instance.According to traits of Chinese information processing, the paper puts forward a model of e-government semantic information automatic extraction and annotation. The model based on Semantic Web and discusses in detail four important issues: Chinese words automatic segmentation and extraction, semantic annotation.At last the paper studies two model involved in Ontology-based e-government information processing issue: a model of mayor mailbox classified automatic processing system; And then a model of ontology-based government semantic portal is presented to make it easy for users to access the right information which they want.更多还原