哈尔滨工程大学学报2011，Vol.32Issue(11)：1484-1488,5.DOI:10.3969/j.issn.1006-7043.2011.11.016
时间序列部分周期模式的更新算法
Partial periodic pattern updating technology in time series databases
摘要
Abstract
In order to solve the problem of high complexity in the computation of the on-line incremental partial periodic pattern mining, this paper presented mining technology from a time series database based on a moving-window. In the process of data mining in time series, in some scenarios, it is only necessary to mine the partial periodic patterns in the recent time series databases in order to forecast the future action trend in time series. Therefore, during the mining, it needs to mine the partial periodic patterns in the recent time series databases using time window based on the former mining results. The proposed incremental on-line mining algorithm can focus the discovery on the recent data using a moving-window, which only needs to scan the data set in the specified time window two times at most. The algorithm was demonstrated in synthetic time series databases and the traffic time series databases. The experimental results show that the new algorithm has a higher computing efficiency than the non-moving-window algorithm in many aspects for large databases.关键词
对间序列/部分周期模式/移动窗/频繁模式Key words
time series/ partial periodic pattern/ moving-window/ frequency pattern分类
信息技术与安全科学引用本文复制引用
王晓晔,肖迎元,张德干..时间序列部分周期模式的更新算法[J].哈尔滨工程大学学报,2011,32(11):1484-1488,5.基金项目
国家自然基金资助项目 (61170174,61001174) （61170174,61001174）
