Concept shift detection for frequent itemsets from sliding windows over data streams

Jia Ling Koh*, Ching Yi Lin

*此作品的通信作者

研究成果: 書貢獻/報告類型會議論文篇章

11 引文 斯高帕斯(Scopus)

摘要

In a mobile business collaboration environment, frequent itemsets analysis will discover the noticeable associated events and data to provide important information of user behaviors. Many algorithms have been proposed for mining frequent itemsets over data streams. However, in many practical situations where the data arrival rate is very high, continuous mining the data sets within a sliding window is unfeasible. For such cases, we propose an approach whereby the data stream is monitored continuously to detect any occurrence of a concept shift. In this context, a "concept-shift" means a significant number of frequent itemsets in the up-to-date sliding window are different from the previously discovered frequent itemsets. Our goal is to detect the notable changes offrequent itemsets according to an estimated changing rate of frequent itemsets without having to perform mining of the frequent itemsets at every time point. Consequently, for saving the computing costs, it is triggered to discover the complete set of new frequent itemsets only when any significant change is observed. The experimental results show that the proposed method detects concept shifts of frequent itemsets both effectively and efficiently.

原文英語
主出版物標題Database Systems for Advanced Applications - DASFAA 2009 International Workshops
主出版物子標題BenchmarX, MCIS, WDPP, PPDA, MBC, PhD
頁面334-348
頁數15
DOIs
出版狀態已發佈 - 2009
事件International Workshops on Database Systems for Advanced Applications, DASFAA 2009: BenchmarX, MCIS, WDPP, PPDA, MBC, PhD - Brisbane, QLD, 澳大利亚
持續時間: 2009 四月 202009 四月 23

出版系列

名字Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
5667 LNCS
ISSN(列印)0302-9743
ISSN(電子)1611-3349

其他

其他International Workshops on Database Systems for Advanced Applications, DASFAA 2009: BenchmarX, MCIS, WDPP, PPDA, MBC, PhD
國家/地區澳大利亚
城市Brisbane, QLD
期間2009/04/202009/04/23

ASJC Scopus subject areas

  • 理論電腦科學
  • 電腦科學(全部)

指紋

深入研究「Concept shift detection for frequent itemsets from sliding windows over data streams」主題。共同形成了獨特的指紋。

引用此