计算机工程与应用 ›› 2013, Vol. 49 ›› Issue (19): 119-122.

• 数据库、数据挖掘、机器学习 • 上一篇    下一篇

面向对象概念格的压缩

陈永平1,杨思春2   

  1. 1.马鞍山职业技术学院 计算机系,安徽 马鞍山 243000
    2.安徽工业大学 计算机学院,安徽 马鞍山 243002
  • 出版日期:2013-10-01 发布日期:2015-04-20

Reduction of object-oriented concept lattices

CHEN Yongping1, YANG Sichun2   

  1. 1.Department of Computer Science, Ma’anshan Technical College, Ma’anshan, Anhui 243000, China
    2.School of Computer Science, Anhui University of Technology, Ma’anshan, Anhui 243002, China
  • Online:2013-10-01 Published:2015-04-20

摘要: 概念格理论是知识处理与分析的一种有力工具,在知识发现和数据挖掘等众多领域有着重要的应用。引入了概念相似度新的计算方法,由对象和属性共同确定概念之间的相似程度,进而产生概念邻域,并根据概念间相似程度来控制概念邻域的大小,删除不必要的节点,从而控制面向对象概念格中节点的个数,实现面向对象概念格的压缩和知识库简化。示例表明,当参数的值较小时,压缩效果明显。

关键词: 形式背景, 概念格, 面向对象概念格, 相似度, 邻居

Abstract: Concept lattice theory is a powerful tool for processing and analysis of knowledge, knowledge discovery and data mining, and other important applications. A new method of similarity calculation of concepts is introduced. Objects and properties are both used to determine the similarity of concepts, generate the concept neighborhood and control its size according to the similarity degree of concepts. And then, it removes unnecessary nodes, to control the number of nodes in the object-oriented concepts, realization of object-oriented concepts simplify the compression and the knowledge base. The examples show that the compressing of object-oriented concept lattice is more effect when parameter values are smaller.

Key words: formal context, concept lattice, object-oriented concept lattice, similarity degree, neighborhood