首页> 外国专利> An ontology based knowledge base construction method using semantic role labeling

An ontology based knowledge base construction method using semantic role labeling

机译:基于语义角色标记的基于本体的知识库构建方法

摘要

The present invention relates to an ontology-based knowledge base construction method for automatically converting a natural language sentence into an ontology sentence. The ontology-based knowledge base construction method comprises: (a) a step of generating an ontology property and case-semantic role bundle connection model (hereinafter referred to as a connection model); a step of receiving a natural language sentence to be converted; (c) a step of choosing a semantic role for each element of the natural language sentence; (d) a step of detecting an ontology property by using a case and the semantic role from the choice of the semantic role and referring to the connection model; (e) a step of performing named entity recognition for the natural language sentence; (f) a step of searching for a resource likely to correspond to a subject or an object of an ontology sentence based on proper nouns extracted through the named entity recognition technique in the (g) step; (g) a step of generating an ontology sentence by combining the ontology property detected in the step (d) and the entities searched in the step (f); and (h) a step of checking the ontology sentence. The ontology-based knowledge base construction method can allow a computer to automatically process knowledge and can be used in various fields by converting vastly accumulated natural language knowledge into more logically rigorous ontology-based knowledge.
机译:基于本体的知识库构建方法技术领域本发明涉及一种基于本体的知识库构建方法,用于将自然语言语句自动转换为本体语句。基于本体的知识库的构建方法包括:(a)生成本体属性和案例语义角色捆绑连接模型(以下称为连接模型)的步骤;接收要转换的自然语言句子的步骤; (c)为自然语言句子的每个元素选择语义角色的步骤; (d)通过从语义角色的选择和引用连接模型中使用案例和语义角色来检测本体属性的步骤; (e)对自然语言句子进行命名实体识别的步骤; (f)在(g)步骤中,基于通过所述命名实体识别技术提取出的专有名词,搜索可能对应于本体论句子的主语或宾语的资源的步骤; (g)通过组合步骤(d)中检测到的本体属性和步骤(f)中搜索到的实体,生成本体语句的步骤; (h)检查本体语句的步骤。基于本体的知识库构建方法可以使计算机自动处理知识,并且可以通过将大量积累的自然语言知识转换为逻辑上更严格的基于本体的知识,将其用于各个领域。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号