×

联系我们

方式一(推荐):点击跳转至留言建议,您的留言将以短信方式发送至管理员,回复更快

方式二:发送邮件至 nktanglan@163.com

学生论文

论文查询结果

返回搜索

论文编号:689 
作者编号:2120062403 
上传时间:2008/6/20 11:45:13 
中文题目:基于Ontology的Web表格数值知识  
英文题目:Research and Realization of Nu  
指导老师:王知津 
中文关键字:Web表格;信息抽取;Ontology学习; 
英文关键字:Web Tables;Information Extract 
中文摘要:表格(Tables)作为一种重要的信息表现形式已广泛地应用于Web页面中,发掘表格中所蕴涵的信息和知识具有重要意义。Web表格信息抽取即是从Web表格中抽取语义一致性的、结构化表示的数据和知识,它在知识管理、信息检索、Web挖掘、摘要提取以及对移动设备的内容传递等应用中有着非常广泛的用途。目前Web表格信息抽取的研究中效果最好的是基于Ontology的方法,但Ontology的构建较为复杂,完全利用手工构建Ontology是一项艰巨的任务。在总结相关研究的基础上,本文提出一种基于Web表格的Ontology学习方法,实现了Ontology的半自动构建。并利用构建的Ontology来指导Web表格信息抽取过程。根据知识管理的需要,从“知识元”层面将抽取结果表示成数值知识元形式,揭示表格个体中的“知识元”与领域共性中的“知识结构”的链接关系。 本文的第一章是绪论,介绍研究的意义和目的、研究内容、研究方法和创新点。 第二章是对目前Web表格信息抽取研究的综述,以明确Web表格信息抽取概念及其过程、关键技术、应用及发展趋势。 第三章简要梳理了知识元、Ontology及Ontology学习的基本理论、研究进展和应用现状。 第四章提出了基于Ontology的Web表格数值知识元抽取系统的总体架构,并对系统各模块进行了设计。 第五章深入阐述了系统关键模块的实现和系统性能的实验分析,并介绍了系统的最终应用。系统实现所涉及的关键技术包括初始Ontology的构建、概念学习技术、关系学习技术、实例学习技术、Web表格定位技术、Web表格结构识别技术、Web表格内容整合和数值知识元抽取技术。 第六章是结论部分,指出了本文研究的不足之处,提供进一步研究的方向。 
英文摘要:Table is a important information sources widely used in Web pages, how to discover the information and knowledge in tables is significant. The technology of information extraction over Web tables extract the data and knowledge which express consistently and structurally, they have widely used in Knowledge Management, Information Retrieve, Web Mining, Abstract extraction and content deliver over PDA. At present the technology based on Ontology has the best effect, but construct a full Ontology is a formidable task for engineer. This paper excogitate a semiautomatic method of Ontology learning based on Web tables and implement it. This paper carries out the course of extraction by the Ontology. According to the needs of Knowledge Management, the results of extraction express as the format of numerical knowledge element for linking to “knowledge structure” at specific domain. Chapter 1 is introduction, presents the meaning and aim, content, method, and innovation of the research. Chapter 2 is a survey of research on information extraction over web tables, showing the concept, course, key technologies, applications and development of it. Chapter 3 presents the theroy, development and applications of knowledge element, Ontology and Ontology learning. Chapter 4 designs the system architecture and modules of numerical knowledge element extraction over Web tables based on the Ontology. Chapter 5 in-depth analyses the implement of key modules, system performance, and makes mention of its applications. The key technologies involve concept learning, relations learning, instance learning, Web table location, Web table structure recognition, Web table interpretation and numerical knowledge element extraction. Chapter 6 is conclusion, points out the problems in current researches, and finally presents a prospect of its fruture.  
查看全文:预览  下载(下载需要进行登录)