论文标题:语义万维网在智能信息检索中的应用研究
Study on Applying Semantic Web to Intelligent Information Retrieve
论文作者 邹景华
论文导师 朱庆生,论文学位 硕士,论文专业 计算机软件与理论
论文单位 重庆大学,点击次数 113,论文页数 65页File Size1155k
2005-03-08论文网 http://www.lw23.com/lunwen_125972087/ 语义万维网;领域本体;智能信息检索;语义检索
Semantic Web; Domain Ontology; Intelligent Information Retrieve; Semantic Retrieve
传统信息检索方式下,由于信息资源缺少统一的语义描述,用户难以查找到与需求相关的资源,难以实现信息资源的语义共享。如何使被管理信息资源具有应用程序可以理解的含义,并根据信息资源所具有的领域知识,实现信息资源的语义检索,提高数字化信息资源的利用率,这些问题是信息检索领域所面临的挑战。语义万维网(Semantic Web)是万维网的发明人Tim Berners-Lee 倡导的下一代万维网,旨在赋予万维网上所有资源唯一的标识,并在资源之间建立起机器可处理的各类语义联系。语义万维网是对当代万维网的扩展和延伸,能够以一种明确的、形式化的方式来表示信息资源,提高异构系统之间的互操作性,促进知识共享。本文通过对传统信息检索的基本原理、技术及工具、信息检索发展现状等方面的分析研究,结合万维网的具体特点,提出了基于语义万维网的智能信息检索体系框架,详细描述了智能信息检索系统的设计思想和检索流程,并对系统模型的合理性进行了论证。本文对智能信息检索模型中所涉及的若干关键技术进行了深入的分析研究,提出了有效的解决方案,为智能检索实验系统的顺利实施奠定了良好的基础,这些关键技术主要包括:领域本体构建、信息资源收集、语义推理和检索结果排序。本文最后在理论分析的基础上,开发了基于语义万维网的智能文献检索实验系统-ISearch (Intelligent Search),该系统以计算机领域的科技文献为智能检索对象,构建了相应的领域本体——文献本体和语义词典本体。通过将各种文献资源向知识本体层映射和语义推理处理,可以充分发掘各文献之间蕴含的关联关系,从而在根本上解决了检索中资源对象语义信息缺乏的问题,满足用户对资源对象语义检索的需求。ISearch 实现相对比较复杂的知识检索和二次检索功能,提供领域资源的增值服务。这些功能都是传统的基于关键字的检索方式所难以实现的,在实践中验证了智能信息检索系统模型的可行性。
According to traditional information retrieval pattern, the information lacks a uniform semantic description. It is hard for user to find more relevant information and to realize the semantic share in information resource. How to assign the managed resource a machine understandable meaning so as to find what users really need, and how to realize the semantic searching by means of the domain knowledge, are challenging tasks for information retrieval. Semantic Web is the next generation Web, which is advocated by Tim Berners-Lee who is a founder of the Web. In Semantic Web, every resource has unique identifier, and there are all of semantic relations between resources. The Semantic Web is an extension of current Web in which information is given well-defined meaning and makes computers and people work in cooperation. Have analyzed the principle, technology, tool and feature of traditional retrievals in Web, the paper proposes a framework of intelligent retrieve based on semantic Web. It also describes design idea and process of the intelligent information retrieval system in detail, and proves in theory the validity of the framework. Aims at key techniques of intelligent retrieve system, the paper gives our solutions which establish the theory basis for implement of an intelligent retrieval system. These key techniques include domain ontology construction, information resource collection, semantic inference,and sort of retrieve result set. Based on our theory analysis, we have successfully developed an intelligent paper search system (ISearch). ISearch selects science papers related to computer domain as experiment objects, and builds related domain ontology (paper ontology and semantic dictionary ontology). Compared with traditional key words matching retrieval system, the system can find the internal relationship among resources, and represent user more relevant resource through the mapping from resource layer to ontology layer and the semantic inference process. What’s more, the system can realize some complicate knowledge query and value-added service. ISearch proves the correctness of intelligent information retrieval framework in practice.
|
