基于领域本体的Web信息抽取  

Web information extraction based on domain ontology

作　　者：杨喜权[1] 孔德冉[1] 施华[1] 孙娜[1] 张野[1] 

机构地区：[1]东北师范大学计算机学院,长春130117

出　　处：《信息技术》2009年第9期69-72,共4页Information Technology

摘　　要：针对Web信息在抽取过程中数据精确率不足的问题,提出了一种基于领域本体的Web信息抽取方法。该方法使得领域本体中概念的层次关系和属性的特点得到了充分利用,本体的构造在抽取过程中逐渐得到完善,从而增强本体对应用领域的描述能力。实验证明文中的方法对Web页面的信息抽取具有较高的效率。Dealing with the data inaccuracy in the process of the extraction,this paper proposes a Web information extraction method based on specific domain ontology.It makes full use of the hierarchical relations and the characteristics of attribution in domain ontology,meanwhile,ontology structure is gradually perfected in the process of the extraction.Thereby the ability to describe ontology in application area has been enhanced.It turns out that it has a higher efficiency in the extraction of Web pages.

关 键 词：Web抽取 领域ONTOLOGY 层次关系 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]