计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (4): 141-145.DOI: 10.3778/j.issn.1002-8331.2011.04.039

• 数据库、信号与信息处理 • 上一篇    下一篇

滑动窗口模型下的概率数据流聚类

程转流1,2,胡为成2   

  1. 1.合肥工业大学 计算机与信息学院,合肥 230009
    2.铜陵学院 信息技术与工程管理研究所,安徽 铜陵 244000
  • 收稿日期:2010-03-16 修回日期:2010-07-26 出版日期:2011-02-01 发布日期:2011-02-01
  • 通讯作者: 程转流

Clustering for probabilistic data stream over sliding windows

CHENG Zhuanliu1,2,HU Weicheng2   

  1. 1.School of Computer and Information,Hefei University of Technology,Hefei 230009,China
    2.Institute of Information Technology & Engineering Management,Tongling College,Tongling,Anhui 244000,China
  • Received:2010-03-16 Revised:2010-07-26 Online:2011-02-01 Published:2011-02-01
  • Contact: CHENG Zhuanliu

摘要: 提出一种基于滑动窗口的概率数据流聚类方法PWStream。PWStream采用聚类特征指数直方图保存最近数据元组的信息摘要,在允许的误差范围内删除过期的数据元组;并针对数据流上概率元组提出强簇、过渡簇和弱簇的概念,设计了一种基于距离和存在概率的簇选择策略,从而可以发现更多的强簇。理论分析和实验结果表明,该方法具有良好的聚类质量和较快的数据处理能力。

关键词: 概率数据流, 聚类, 滑动窗口, 直方图

Abstract: An effective clustering algorithm called PWStream for probabilistic data stream over sliding window is developed.The algorithm uses exponential histogram of cluster feature to store the summary information of the most recently arrived tuples,and outdated information is deleted within a certain guaranteed range of error.For the uncertain tuples in data stream,the concepts of strong cluster,transitional cluster and weak cluster are proposed in the PWStream.With these concepts,an effective strategy of choosing cluster based on distance and existence probability is designed,which can find more strong clusters.Theoretical analysis and comprehensive experimental results demonstrate that the proposed method is of high quality and fast processing rate.

Key words: probabilistic data stream, clustering, sliding window, histogram

中图分类号: