信息抽取中领域本体的设计和实现

Design and Implementation of Domain Ontology for Information Extraction

  • 摘要: 领域本体在信息抽取系统有着重要作用。该文介绍了本体、领域本体的基本概念,并探讨了领域本体与信息抽取的关系。提出了信息抽取系统中领域本体的设计准则和实施步骤,实施步骤包括领域本体需求分析、收集本体信息、构建领域本体框架、形式化编码、确认和评价等。在信息抽取原型系统中实现了显示器领域本体,并将该领域本体应用到信息抽取中的命名实体识别、抽取模式获取和主题概念提取等任务中,应用结果表明该方法、步骤是可行的。

     

    Abstract: Domain ontology plays an important role in the system of information extraction. In this paper, after a discuss of the relation of domain ontology and information extraction, the design principles and steps of domain ontology in the system of information extraction are proposed. The domain ontology about monitor is implemeneted in a prototype system of information extraction, and the domain ontology is applied to some tasks of information extraction, including named entity recognition, extraction pattern acquisition, and thematic concept extraction. The application results show that the principles and steps are feasible.

     

/

返回文章
返回