摘要
聚类是数据挖掘中的一个非常活跃的研究领域,聚类的目的就是把数据集分成不同的类,类内相似度高,类间相异度大。本文介绍了在聚类过程中经常遇到的数据结构、变量类型和聚类方法,提出了基于万有引力定律的聚类方法,使聚类的速度和效果有了进一步的提高。
Cluster is one of the most hot research fields of Data Mining. The purpose of cluster is that makes the data set into several clusters. The data objects belong to the same cluster are similar to one another and dissimilar to the ones in other clusters. In this paper, the author introduces the data structures, data variables, and the methods of cluster that we may meet. The cluster method based on gravity can accelerate cluster progress and make a better effort.
出处
《安阳工学院学报》
2006年第4期40-43,共4页
Journal of Anyang Institute of Technology
关键词
数据挖掘
聚类
万有引力定律
重心
Data Mining, cluster, gravity theory, center of gravity
