文献类型：期刊文献

中文题名：New Clustering Method in High-Di mensional Space Based on Hypergraph-Models

英文题名：New clustering method in high-dimensional space based on hypergraph-models

作者：Chen, Jian-Bin[1,2]; Wang, Shu-Jing[3]; Song, Han-Tao[1]

第一作者：陈建斌;Chen, Jian-Bin

通讯作者：Chen, J.-B.

机构：[1] School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081, China; [2] Business College, Beijing Union University, Beijing 100025, China; [3] China Aviation Accounting Center, Beijing 100028, China

第一机构：School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081, China

通讯机构：[1]School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081, China

年份：2006

卷号：15

期号：2

起止页码：156-161

中文期刊名：北京理工大学学报：英文版

外文期刊名：Journal of Beijing Institute of Technology (English Edition)

收录：EI(收录号：20063710111423);Scopus(收录号：2-s2.0-33748471324)

语种：英文

中文关键词：高维聚类;超图模型;数据挖掘;传统聚类算法

外文关键词：Algorithms - Data mining - Dimensional stability - Evaluation - Graph theory - Mathematical models

摘要：To overcome the limitation of the traditional clustering algorithms which fail to produce meaningful clusters in high-dimensional, sparseness and binary value data sets, a new method based on hypergraph model is proposed. The hypergraph model maps the relationship present in the original data in high dimensional space into a hypergraph. A hyperedge represents the similarity of attribute-value distribution between two points. A hypergraph partitioning algorithm is used to find a partitioning of the vertices such that the corresponding data items in each partition are highly related and the weight of the hyperedges cut by the partitioning is minimized. The quality of the clustering result can be evaluated by applying the intra-cluster singularity value. Analysis and experimental results have demonstrated that this approach is applicable and effective in wide ranging scheme.
To overcome the limitation of the traditional clustering algorithms which fail to produce meaningful clusters in high-dimensional, sparseness and binary value data sets, a method based on hypergraph model is proposed. The hypergraph model maps the relationship present in the original data in high dimensional space into a hypergraph. A hyperedge represents the similarity of attribute-value distribution between two points. A hypergraph partitioning algorithm is used to find a partitioning of the vertices such that the corresponding data items in each partition are highly related and the weight of the hyperedges cut by the partitioning is minimized. The quality of the clustering result can be evaluated by applying the intra-cluster singularity value. Analysis and experimental results have demonstrated that this approach is applicable and effective in wide ranging scheme.