• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2011, Vol. 33 ›› Issue (10): 149-153.

• 论文 • 上一篇    下一篇

基于自适应网格划分的数据流聚类算法

邢长征,张国凯   

  1. (辽宁工程技术大学电子与信息工程学院,辽宁 葫芦岛 125105)
  • 收稿日期:2011-03-15 修回日期:2011-05-28 出版日期:2011-10-25 发布日期:2011-10-25

A Data Stream Clustering Algorithm Based on Adapative Grid Partitioning

XING Changzheng,ZHANG Guokai   

  1. (School of Eletronics and Information Engineering,Liaoning Technical University,Huludao 125105,China)
  • Received:2011-03-15 Revised:2011-05-28 Online:2011-10-25 Published:2011-10-25

摘要:

本文提出了一种基于自适应网格划分的数据流聚类算法。通过采用网格的自适应划分,对传统的基于密度网格的数据流聚类算法,以均衡划分网格的方法进行改进,使网格的划分更加合理,减少硬性划分对结果可能造成的影响,提高了硬性划分边界的精度。同时采用剪枝方法,减少了算法的执行时间。最后,通过实验验证了该算法的有效性。

关键词: 数据流, 聚类, 滑动窗口, 网格

Abstract:

This paper proposes a data stream clustering algorithm based on adaptive grid partitioning.By using adaptive grid partitioning to improve the traditional methed of dividing grids in a balanced method,we make  the grid division more reasonable and reduce the impact on the result, which improves the precision of grid partitioning.Using a pruning method to ruduce the algorithm’s execution time is effective.Finally, the experimental results verify the effectiveness of the proposed algorithm.

Key words: data stream;clustering;sliding window;grid