计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (21): 1-4.DOI: 10.3778/j.issn.1002-8331.2008.21.001

• 博士论坛 • 上一篇    下一篇

数据流突发检测研究与进展

袁志坚,王 乐,田 李,贾 焰,杨树强   

  1. 国防科学技术大学 计算机学院,长沙 410073
  • 收稿日期:2008-03-27 修回日期:2008-05-06 出版日期:2008-07-21 发布日期:2008-07-21
  • 通讯作者: 袁志坚

Research and development of burst detection in data streams

YUAN Zhi-jian,WANG Le,TIAN Li,JIA Yan,YANG Shu-qiang   

  1. School of Computer,National University of Defense Technology,Changsha 410073,China
  • Received:2008-03-27 Revised:2008-05-06 Online:2008-07-21 Published:2008-07-21
  • Contact: YUAN Zhi-jian

摘要: 数据流是不断变化且难以预测的。因此,在数据流中进行突发检测,是数据流内在的,固有的问题之一。所谓突发,指的是特定时间段内的数据量显著异常于其它时间段。如何实时地相对精确地检测出数据流中的突发并良好地呈现给用户,国内外已展开相关研究,并成为数据流挖掘领域的热点问题之一。论文综述国内外数据流突发检测的研究现状,归纳与分析现有研究工作的适用场景,并给出研究的焦点及热点,最后展望了该领域的前景。

关键词: 数据流, 突发检测, 弹性窗口, 滑动窗口

Abstract: Burst detection is an inherent problem for data streams,so it has attracted extensive attention in research community due to its broad applications.Burst on data stream means abnormal number of data emerges during current period of time comparing with last period of time.How to detect burst in data stream real-timely and accurately,which has been a hotspot in stream mining domain.This paper summarizes current research works in burst detection,sums up and analyzes the applicability of existing works and gives the focus of these researches.Finally the future in this area is given.

Key words: data stream, burst detection, elastic window, sliding window