基于HNC的命题逻辑推理方法  

Approach for Propositional Logic Reasoning Based on HNC

作　　者：李颖[1] 田永昌[2] 

机构地区：[1]装甲兵工程学院科研部,北京100072 [2]装甲兵工程学院研究生管理大队,北京100072

出　　处：《装甲兵工程学院学报》2013年第2期70-74,共5页Journal of Academy of Armored Force Engineering

摘　　要：针对面向自然语言推理中计算机难以准确提取命题并进行符号化的难题,运用HNC(Hieratical Network of Concepts)理论中的语言逻辑概念与句类分析技术,从概念的角度出发,对自然语言描述的推理问题进行理解处理,使计算机能自动提取出推理条件中蕴涵的命题,然后将命题进行公式化处理,最后得到用自然语言描述的结论命题。试验结果表明:该方法能准确理解自然语言中的语义逻辑关系,完成命题逻辑推理。To overcome the difficulty for computer-based extraction and formulization in natural language oriented reasoning,language logic concept and sentence category analysis technology from HNC(Hieratical Network of Concepts) are adopted to deal with the reasoning problem described in natural language.The paper first discusses automatic extraction of underlying propositions,which are later formulized for ultimate conclusion described in natural language.Test results illustrate the effectiveness of proposed approach in understanding underlying semantic logic relation in natural language and completing propositional logic reasoning.

关 键 词：自然语言理解 句类分析 命题 推理 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]