计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (23): 151-153.DOI: 10.3778/j.issn.1002-8331.2008.23.046

• 数据库、信号与信息处理 • 上一篇    下一篇

基于CA模型的凝固聚类算法

张俊溪,薛惠锋,苏锦旗   

  1. 西北工业大学 自动化学院,西安 710072
  • 收稿日期:2007-10-16 修回日期:2008-01-21 出版日期:2008-08-11 发布日期:2008-08-11
  • 通讯作者: 张俊溪

Agglomerative clustering algorithm based on CA

ZHANG Jun-xi,XUE Hui-feng,SU Jin-qi   

  1. College of Automatic Control,Northwestern Polytechnical University,Xi’an 710072,China
  • Received:2007-10-16 Revised:2008-01-21 Online:2008-08-11 Published:2008-08-11
  • Contact: ZHANG Jun-xi

摘要: 聚类是数据挖掘领域的重要研究内容之一。参考基于元胞自动机距离变换算法模型,构建了基于CA模型的凝固聚类算法,该算法在CA模型演化的过程中,可以产生完整的层次聚类结果,同时对簇间的距离实现了度量,能够处理形状复杂的聚类对象,具有较好的向高维空间的推广能力以及并行计算的特性。最后通过两组聚类数据进行了实证研究,验证了该算法的有效性。

关键词: 数据挖掘, 聚类, 元胞自动机, 并行计算, 凝固聚类

Abstract: Clustering is one of the most important research in data mining area.Referring to the distance transform method based on Cellular Automata(CA),an agglomerative clustering algorithm is constructed based on CA.The algorithm can generate whole cluster results and the measurement between clusters during the evolution process of CA.It can also treat with non-convex set of objects.So it has the characteristics of extending to high dimensional space and parallel computing.Simulation results of two groups of sample are given to illustrate the effectiveness of the algorithm.

Key words: data mining, clustering, Cellular Automata(CA), parallel computing, agglomerative clustering