摘要
随着分类属性数据集的应用越来越广泛,获取含有分类属性数据集的聚类边界的需求也越来越迫切。为了获取聚类的边界,在定义分类数据的边界度和聚类边界的基础上,提出了一种带分类属性数据的聚类边界检测算法——CBORDER。该算法首先利用随机分配初始聚类中心和边界度对类进行划分并获取记录边界点的证据,然后运用证据积累的思想多次执行该过程来获取聚类的边界。实验结果表明,CBORDER算法能有效地检测出高维分类属性数据集中聚类的边界。
With the wide application of categorical-attribute dataset,the demand of obtaining the cluster boundary of categorical-attribute dataset becomes more and more urgent.In order to get cluster boundaries,a categorical-attribute data boundary detection algorithm： CBORDER（Categorical dataset BORDER detection algorithm） was proposed.In this algorithm,firstly,this paper initialized the center of cluster by using random allocation and utilizing boundary-degree to partition clusters;at the same time,the evidence of captured boundary records was got.Then,based on the evidence accumulation,the above procedure was executed repeatedly to acquire the boundaries of clusters at the end.The experimental results demonstrate that CBORDER can effectively detect the boundaries of the high-dimensional categorical data.
出处
《计算机应用》
CSCD
北大核心
2012年第6期1654-1656,1669,共4页
journal of Computer Applications
基金
河南省重点科技攻关项目(112102310073)
河南省教育厅自然科学研究计划项目(2009A520028)
关键词
边界度
证据积累
聚类边界
分类数据
boundary-degree
evidence accumulation
cluster boundary
categorical data
