基于场论的本体中概念相似度求解方法

发布时间：2018-02-27 18:34

本文关键词： 本体概念相似度场论语义场语义引力元语义距离　出处：《计算机集成制造系统》2016年03期 　论文类型：期刊论文

【摘要】：针对目前本体中概念相似度求解方法孤立考虑概念信息量或概念关系的问题,提出基于场论的概念相似度求解方法。该方法首先明确概念的语义关系是由概念包含的信息决定且隐性的概念关系通过显性的概念关系推出,并将概念包含的信息划分为共性信息与特性信息两类。结合场论的原理和方法,将本体映射为与物理场具有相同特性的语义场,将概念的共性信息和特性信息分别转化为语义引力元和语义距离,采用引力场模型计算概念的相似度。结合坦克装甲车辆和舰船领域知识检索需求,建立基于本体的语义检索系统,并通过与五种典型的相似度求解方法进行对比分析,验证了该方法的正确性和有效性。
[Abstract]:To solve the problem of concept information quantity or concept relation in ontology, the concept similarity method is considered in this paper. A conceptual similarity solution based on field theory is proposed. Firstly, it is clear that the semantic relationship of the concept is determined by the information contained in the concept, and the implicit concept relationship is derived from the explicit concept relationship. And the information contained in the concept is divided into two categories: common information and characteristic information. Combined with the principle and method of field theory, ontology is mapped to semantic field with the same characteristics as physical field. The general information and characteristic information of the concept are transformed into semantic gravitational element and semantic distance respectively. The similarity of the concept is calculated by using the gravitational field model, and the requirement of searching knowledge in the field of tank armored vehicle and ship is combined. A semantic retrieval system based on ontology is established and compared with five typical similarity solving methods to verify the correctness and effectiveness of the proposed method.
【作者单位】：北京理工大学机械与车辆学院;中国北方车辆研究所;
【基金】：国家自然科学基金资助项目(51105039) 国防基础科研资助项目(A2220110001)~~
【分类号】：TP391.1

【相似文献】