基于ontology的中文信息抽取系统的研究与实现

【作者】 孟令谦；

【导师】 左志宏；

【作者基本信息】 电子科技大学 ， 计算机软件与理论， 2004， 硕士

【摘要】 本文首先介绍了信息抽取技术及其产生背景和发展历史，分析了信息抽取系统体系结构、信息抽取的分类及信息抽取的关键技术和衡量指标。本文还介绍了ontology的基本知识。在此基础上，本文提出了一种基于ontology的信息抽取新方法。在本方法中，我们把信息抽取和ontology结合起来，先利用领域ontology里的概念、关系、关键字等自动生成抽取规则(Rule)，然后对文章，句子进行语法分析预处理，再利用语法分析的结果和先前生成的抽取规则一起对文档进行信息抽取，最后把抽取的结果以记录的形式输出。 在本文中，我们根据上述方法并结合工程实际情况，设计了一个基于ontology的信息抽取系统并编写代码实现了这个系统。因此在本文中，我们详细介绍了系统的总体框架，系统的各主要模块的设计等。由于本系统是根据规则进行信息抽取的，所以本文还着重说明了一种基于Lex的信息抽取规则，介绍了它们的语法、抽取信息的方法以及自动产生抽取规则的算法等。 我们还详细介绍了本系统的实现方法，包括各个主要部分的数据结构，流程图等。最后利用此系统对一些样本进行抽取后的结果并对结果进行了分析。更多还原

【Abstract】 In this paper, we first introduce the Information Extraction technology and its developing background and history. We analyse the system architecture, the taxonomy of Information Extraction and the key technology and weighing measure of Information Extraction. And we also introduce the basic knowledge of ontology. Based on this, we present a new approach to extracting information from normal document based on an application ontology that describes a domain of interest. In our approach we combine the Information Extraction with ontology. We first use the concepts, relations and keywords of domain ontology to generate Information Extraction rule automatically and then do grammar parsing on the document. After that we use the result of grammar parsing and Information Extraction rule to do information extraction on document and at last output the result as a list of records.In this paper, according to the approach and engineering reality condition, we designed an Ontology based Information Extracton System and wrote some codes and implemented the system, so in this paper, we introduce the main frame and the designing method of main modals in detail. Because we use the rule to extract information, so we focus on introducing a kind of rule, Lex. We not only introduce the grammar of these rules but also bring out the arithmetic of generating rule automatically.We also introduce the way we implement the system which includes data structure, flow chart etc. Then, we show the user interface of this extraction system and the result which we got from the processing of this system using some test documents. At last we analyse the extraction result.更多还原