基于概念格的概念相似度计算  

Concept Similarity Based on Concept Lattice

作　　者：智慧来[1] 智东杰[2] 刘宗田[1] 

机构地区：[1]上海大学计算机工程与科学学院,上海200072 [2]河南理工大学计算机科学与技术学院,焦作454150

出　　处：《计算机科学》2008年第9期156-157,167,共3页Computer Science

基　　金：国家自然科学基金(项目批准号60575035)

摘　　要：为了提高信息检索的查全率和查准率,经常要处理相似的概念,因此计算概念间的相似度是必要的。概念由对象和属性两部分组成,所以基于概念格相似度计算的也分为两部分:一是利用概念格的层次关系计算对象的相似度,另一部分计算属性的相似度。其次,概念间的相似度与概念在概念格中的深度有关,计算时利用深度对结果加以修正。计算模型利用了概念较为完整的信息,计算结果与人工判断基本吻合。In order to improve the recall and precision, similar concepts are often processed during the information retrieval. With this regard, the possibility of evaluating concept similarity is acquiring an increasing relevance, since it allows the identification of different concepts that are semantically close. For every concept consist of two parts which include objects and attributes, so the computing model also has two parts. First part deals with the objects similarity by using hierarchy structure of concept lattice. The second part deals with the attributes similarity by using the expert＇s knowledge. Beside this, the depth also has relevance with concepts similarity. So the result is altered with concepts＇ depth. This model uses concept＇s integral information, and the result fits the judge given by expert.

关 键 词：概念格 概念相似度 对象 属性 深度 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构] TP311.13[自动化与计算机技术—计算机科学与技术]