首页 | 本学科首页   官方微博 | 高级检索  
     

一种多聚类中心的划分方法
引用本文:宁永恒,杨小兵,王康健.一种多聚类中心的划分方法[J].中国计量学院学报,2014(2):204-208.
作者姓名:宁永恒  杨小兵  王康健
作者单位:中国计量学院信息工程学院,浙江杭州310018
基金项目:国家自然科学基金资助项目(No.61100160).
摘    要:提出了一种多聚类中心的算法(DMC),新算法将原划分算法中每类单个中心替代为多个中心,判断数据归属时以最近中心为目标进行划分,最后对多个中心调整得到聚类结果.实验证明,在不增加算法复杂度的前提下,DMC能够客服划分方法倾向识别大小相近、凸形分布的数据的缺陷,解决不同数据分布的问题,获得良好的聚类效果.

关 键 词:划分方法  多聚类中心  不同数据形状数据  时间复杂度

A partitioning method of multicenters
NING Yongheng,YANG Xiaobing,WANG Kangjian.A partitioning method of multicenters[J].Journal of China Jiliang University,2014(2):204-208.
Authors:NING Yongheng  YANG Xiaobing  WANG Kangjian
Affiliation:(College of Information Engineering, China Jiliang University, Hangzhou 310018, China)
Abstract:A multicenter partitioning clustering algorithm(DMC) was proposed.In DMC,the single center was substituted with the multicenter.The shortest center was defined as the partition goal,and the cluster result was obtained by adjusting the multicenter.Experimental results showed that DMC could overcome the shortcomings of the tendency to identify similar sizes and the convex distribution of data in the partitioning method.It can solve the problem of different data distribution and can obtain a better result without increasing complexity.
Keywords:partitioning method  multiple partitioning centers  time complex  data of uninform size
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号