The Study of Parallel K-Means Algorithm

Author

Zhang, Yufang ; Xiong, Zhongyang ; Mao, Jiali ; Ou, Ling

Author_Institution

Dept. of Comput. Sci., Chongqing Univ.

Volume

2

fYear

0

fDate

0-0 0

Firstpage

5868

Lastpage

5871

Abstract

Clustering analysis plays an important role in scientific research and commercial application. K-means algorithm is a widely used partition method in clustering. As the dataset´s scale increases rapidly, it is difficult to use k-means to deal with massive amount of data. A parallel strategy is incorporated into clustering method and a parallel k-means algorithm is proposed. For enhancing the efficiency of parallel k-means, dynamic load balance is introduced. Data parallel strategy and master/slave model are adopted. The experiments demonstrate that the parallel K-means has higher efficiency and universal use

Keywords

parallel algorithms; pattern clustering; resource allocation; clustering analysis; data parallel strategy; dynamic load balancing; master/slave model; parallel k-means algorithm; partition method; Clustering algorithms; Concurrent computing; Data mining; Distributed computing; Master-slave; Parallel algorithms; Parallel architectures; Parallel processing; Parallel programming; Partitioning algorithms; PVM; Parallel Clustering; the K-means Algorithm;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Control and Automation, 2006. WCICA 2006. The Sixth World Congress on

Conference_Location

Dalian

Print_ISBN

1-4244-0332-4

Type

conf

DOI

10.1109/WCICA.2006.1714203

Filename

1714203