用基于移动均值的索引实现时间序列相似查询 - 厦门大学数据库实验室.pdf

ISSN 1000-9825, CODEN RUXUEW E-mail: jos@ Journal of Software , Vol.19, No.9, September 2008, pp.2349−2361 DOI: 10.3724/SP.J.1001.2008.02349 Tel/Fax: +86-10 © 2008 by Journal of Software . All rights reserved. ∗ 用基于移动均值的索引实现时间序列相似查询 1,2+ 1,2 1,2 林子雨 , 杨冬青 , 王腾蛟 1(高可信软件技术教育部重点实验室(北京大学),北京 100871) 2(北京大学 信息科学技术学院,北京 100871) Similarity Search of Time Series with Moving Average Based Indexing LIN Zi-Yu1,2+, YANG Dong-Qing1,2, WANG Teng-Jiao1,2 1(Key Laboratory of High Confidence Software Technologies (Peking University), Ministry of Education, Beijing 100871, China) 2(School of Electronics Engineering and Computer Science, Peking University, Beijing 100871, China) + Corresponding author: E-mail: cainiu@263.net Lin ZY, Yang DQ, Wang TJ. Similarity search of time series with moving average based indexing. Journal of Software, 2008,19(9):2349−2361. /1000-9825/19/2349.htm Abstract : In this paper, a new method, called MABI (Moving Average Based Indexing) is proposed to effectively deal with the issue of ε-search query in subsequence matching. Two important theorems, distance reduction theorem and DRR relation theorem, are proposed here to be the basis of MABI . DRR relation theorem has strong capability in “pruning” those unqualified candidate sequences so as to achieve the purpose of fast similarity search. Furthermore, in order to speed up the process of similarity search, a multi-way balanced tree structure, called BATON* proposed by Jagadish et al., is also introduced here, with some modification, to construct the index from time series. Extensive experiments over a stock excha