基于语义的专业文献学习系统研究与实现

论文字数:27995,页数:46  有开题报告，任务书

摘要
 文献学习是整个学习过程中的一个重要环节，在进一步深造学习中更是不可或缺的。注重文献学习，不仅可以少走弯路，巩固专业知识，而且还可以结合前辈的经验在一定程度上进行创新。并且，最前沿的理论也都是在文献中体现的。所以，准确快速找到所需要的专业文献可以提高学习效率和提升专业素养。但是目前的文献检索基本是通过文字语法匹配和全文检索技术来实现的，检索的结果会出现漏检或不相关文献，学习者花费大量时间来检索还有可能找不到所需的文献。
 本文通过对语义万维网结构的分析、本体知识的学习以及RDF的研究，提出基于语义的专业文献学习系统框架。该框架旨在提取专业文献的语义信息，将专业文献分类管理，实现对专业文献的语义分析，赋予查询系统足够的语义信息，以解决当前文献检索系统中同义词难以识别、相关查找困难等问题，便于进一步学习和交流。
 文献的元数据描述、领域本体的构建及RDF语义查询是本文研究的重点，也是进行语义检索的理论基础。首先，本文针对专业文献结构和学习者检索中所关心的主要内容，建立了文献的元数据库，然后通过对知识结构和特征的分析，针对计算机网络课程的领域知识，构建了一个领域本体。通过领域本体与元数据之间的映射关系将知识点与具体的文献联系起来，得到相关的知识结构，然后添加领域资源的RDF描述，通过RDF的描述方法和蕴涵规则进行有限形式的推理。
 最后开发了一个基于语义的专业文献学习实验系统。该系统以计算机网络领域的知识和学习资源为检索对象，尝试解决查询中同义词无法识别，相关信息查找困难等问题。实验证明了该框架信息资源的语义信息得到充分的体现以及语义检索策略的可行性。

关键词：语义万维网，本体，资源描述框架，语义检索，元数据
Research and Realization on Semantic Based Specialized Literatures Learning System
Abstract
 The literature study is important in an entire study process, and also indispensable in the further study. To attention literature study, not only walk little tortuous path, consolidated specialized knowledge, moreover also unify senior''s experience to carry on the innovation in the certain degree. And, the most front theories are also manifested in the literature. Therefore, the specialized literatures which are found quickly and accurately may enhance the study efficiency and promote specialized accomplishment. But nowadays literature retrieval is basic on matching grammar and full text retrieval technology. The result is not satisfied to learners. It takes lots of time to search articles, but maybe they are not in need.
 This paper proposes the framework of specialized literature study system through analyzing the semantic web structure, studying ontology knowledge, and researching RDF inference technology. The purpose of the framework is to withdraw the semantic information of the specialized literature, and to manage the specialized literature classification of files. It realizes semantic analysis of specialized literature and gives the inquiry system enough semantic information to solve problems that current literature retrieval system is difficult to distinguish the synonym words and search correlation information and so on, in order to study further and communicate each other.
 The keynotes of this article are metadata description for literature, subject ontology creation, RDF semantic retrieval. They are also the theory bases of semantic retrieval. Firstly, according to the structure of specialized literatures and main content which students care for, we build the metadata database for literatures. Then after analyzing knowledge structure and characteristics, we build a subject ontology which is directed against the subject knowledge of computer network course. According to representation relationship between the field ontology and metadata, the specialized literature can related to its subject knowledge feature and obtain knowledge structure. And then add the semantic description about resources by RDF, it support some limited inference according to RDF description method and entailment rules.
 We have developed a specialized literature study system on computer network field based on semantic. This system tries to solve problems that ordinary systems can not do, such as identify synonyms, relative retrieval. The result of experimentation proves that this retrieval framework makes semantic of specialized literatures sufficiently emerged and the good feasibility of search strategy.
 
Key Word：Semantic Web, Ontology, RDF, Semantic Retrieval, Metadata
目录
1. 绪论 1
1.1 研究背景及目的 1
1.2 国内外研究现状 2
1.3 本文主要工作 3
2. 语义万维网 4
2.1 语义web技术 4
2.1.1 语义web概念 4
2.1.2 语义web结构 5
2.1.3 语义web的基础和核心 10
2.2 Ontology的定义和内涵 11
2.3 元数据 13
2.3.1 元数据概念 13
2.3.2 学习对象元数据标准 14
2.3.3 元数据结构 15
2.4 资源描述框架(RDF) 15
2.4.1 RDF概述 15
2.4.2 RDF与XML 16
2.4.3 RDF特性 18
3. 语义检索 20
3.1 搜索引擎发展趋势 20
3.2 语义检索的实现思路和方法 21
3.3 工具介绍 22
3.3.1 建立本体工具Protégé 22
3.3.2 Jena在语义检索中的作用 23
4. 系统设计 26
4.1 需求分析 26
4.2 设计目标 27
4.3 开发工具与平台 27
4.4 总体设计 27
4.4.1 系统流程 27
4.4.2 设计拟用途径 28
4.4.3 领域本体 29
4.4.4 领域规则库 30
4.4.5 解析推理 31
4.5 系统实现 33
4.5.1 系统界面 33
4.5.2 实验结果 34
5. 结论与展望 35
5.1 结论 35
5.2 展望 35
致谢 37
参考文献 38