采用改进的属性拓扑图探索形式背景属性约简之方法

发布时间：2018-01-08 04:13

本文关键词：采用改进的属性拓扑图探索形式背景属性约简之方法　出处：《小型微型计算机系统》2017年12期 　论文类型：期刊论文

【摘要】：形式背景的属性约简可以使概念格的结构更加简便,从而有利于数据的规则提取.首先,从图论的角度研究形式背景的属性约简问题,即结合有向图的性质,提出了改进的属性拓扑图,并给出了在此图下的关联矩阵.其次,通过对属性拓扑图的分析,给出了净化形式背景中的关于概念和交不可约元的判断定理.进而,利用概念中属性特征和交不可约元之间的关系,提出了不必要属性的判别定理,并在此基础之上,给出了求形式背景属性约简的算法.最后,实例分析表明新的属性约简方法在一定程度上提高了概念格的构造效率,从而验证了该算法的可行性和有效性.
[Abstract]:Attribute reduction of formal background can make the structure of concept lattice more convenient, which is beneficial to the rule extraction of data. Firstly, the attribute reduction problem of formal background is studied from the view of graph theory, that is, combining the properties of directed graph. In this paper, an improved attribute topology graph is proposed, and the correlation matrix under the graph is given. Secondly, the analysis of the attribute topology graph is given. The judgment theorems about concepts and intersection irreducible elements in the background of purifying form are given. Furthermore, by using the relationship between attribute characteristics and intersection irreducible elements in the concept, the judgement theorem of unnecessary attributes is put forward. On this basis, the algorithm of finding formal background attribute reduction is given. Finally, the case study shows that the new attribute reduction method improves the construction efficiency of concept lattice to a certain extent. The feasibility and effectiveness of the algorithm are verified.
【作者单位】：河北大学数学与信息科学学院;
【基金】：国家自然科学基金项目(61572011)资助河北省自然科学基金项目(A2013201119)资助
【分类号】：TP18
【正文快照】： 1引言自1982年Wille[1]以序理论和完备格理论为基础,面向概念发现以及层次结构设计的概念格理论,作为数据分析的有效工具已被广泛应用于软件工程[2]、知识发现[3]、信息检索等领域[4,5].概念是知识表达的一种手段,数据库中知识发现的过程就是将数据库中蕴含的知识形式化成有用

【相似文献】