QQ客服 官方微博 反馈留言

	大规模时间序列数据库降维及相似搜索
	李爱国 ; 覃征 ; LI Ai-Guo ; QIN Zheng
	2010-06-09 ; 2010-06-09
关键词	数据库 时间序列 相似搜索 数据挖掘 查询 database time series similarity search data mining query TP311.13
其他题名	Dimensionality Reduction and Similarity Search in Large Time Series Databases
中文摘要	提出一种基于分段多项式表示(PPR)的时间序列数据库相似查询的系统化方法.PPR是一类基于线性多项式回归的正交变换.用PPR变换索引时间序列数据在理论上具备非漏报性质.文中分析了PPR的计算复杂性以及查询阈值的下界,并提出了一种衡量时间序列相似查询算法之查询效率的定量指标.与基于离散傅立叶变换(DFT)和离散小波变换(DWT)的时间序列相似查询算法所作的对比实验表明,所提算法可以用低的索引结构维数获得高的查询效率.; The problem of similarity search in time series databases has attracted much research interest in the database and data mining communities in the last decade. A systemic method of indexing and similarity searching in time series databases based on Piecewise Polynomial Representation (PPR) is proposed in this paper. The idea is to map each sub-sequence into a small set of multidimensional rectangles in feature space that is spanned by base of linear polynomial. PPR is a linear polynomial representation, and PAA (Piecewise Aggregate Approximation), an well known time series compression technique, is a special case of PPR. PPR is used as an efficient dimensionality reduction technique to permit similarity search over large time series databases without false dismissals. Computational complexity of PPR is O(n). The lower boundaries of search threshold are estimated, and a detailed performance anlysis of proposed method is presented. The experimental results demonstrate that performances of proposed method are superior to that of DFT (Discrete Fourier Transform) and DWT (Discrete Wavelet Transform) based index techniques.; 陕西省科学技术发展计划“十五”攻关项目基金(2000K08G12)资助
语种	中文 ; 中文
内容类型	期刊论文
源URL	[http://hdl.handle.net/123456789/53532]
专题	清华大学
推荐引用方式 GB/T 7714	李爱国,覃征,LI Ai-Guo,等. 大规模时间序列数据库降维及相似搜索[J],2010, 2010.
APA	李爱国,覃征,LI Ai-Guo,&QIN Zheng.(2010).大规模时间序列数据库降维及相似搜索..
MLA	李爱国,et al."大规模时间序列数据库降维及相似搜索".(2010).

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。