基于弧度距离的时间序列相似度量

丁永伟; 杨小虎; 陈根才; KavsAJ

doi:10.3724/SP.J.1146.2010.00136

基于弧度距离的时间序列相似度量

doi: 10.3724/SP.J.1146.2010.00136

丁永伟^{*① 杨小虎,},
杨小虎,
陈根才,
KavsAJ

计量
- 文章访问数: 4209
- HTML全文浏览量: 112
- PDF下载量: 1674
- 被引次数: 0
出版历程
- 收稿日期: 2010-02-05
- 修回日期: 2010-07-26
- 刊出日期: 2011-01-19

Radian-distance Based Time Series Similarity Measurement

摘要

摘要: 时间序列的近似表示和相似度量是时间序列数据挖掘的重要任务之一，是进行相似匹配的关键。该文针对现有的各种基于分段线性表示(Piecewise Linear Representation，PLR)相似度量方法存在的序列长度依赖和多分辨率条件下的潜在识别误差等缺点，提出了一种序列分段线性弧度表示和基于弧度距离的相似度量方法，实现了序列的快速在线分割和相似度计算。该方法简洁直观，利用分段弧度对分段趋势进行细粒度划分来保留序列主要形态特征，有效地提高了度量结果的准确性和多分辨率条件下的稳定性。该方法具有序列分割算法独立性特点，可用于时间序列的相似查询、模式匹配、分类和聚类。
- 时间序列 /
- 分段线性表示 /
- 分段趋势 /
- 弧度距离 /
- 相似性
Abstract: Time series approximation representation and similarity measurement is one of the fundamental tasks in time series data mining, and the key to similarity matching. In view of shortcomings of various existing PLR (Piecewise Linear Representation) based similarity measure approaches, like series-length dependent issue and potential recognition error under multi-resolution, a radian based time series piecewise linear representation and radian-distance based similarity measurement are presented to cater for the rapid online segmentation and similarity calculation in this paper. The proposed method is really simple but intuitive, it retains major shape features of the series by using segment radian for fine grained division, and effectively improves the accuracy and reliability of the measurement under multi-resolution. This method is segmentation algorithm independent and can be further applied to similarity query, pattern matching, classification and clustering for time series.
- Time series /
- Piecewise Linear Representation (PLR) /
- Segment trend /
- Radian-distance /
- Similarity

HTML全文

参考文献(1)

Aigner W, Miksch S, and Mller W, et al.. Visual methods for analyzing time-oriented data[J].IEEE Transactions on Visualization and Computer Graphics (TVCG.2008, 14(1):47-60[2]Fu T, Chung F, and Luk R, et al.. Stock time series pattern matching: template-based vsrule-based approaches[J].. Engineering Applications of Artificial Intelligence.2007, 20(3):347-364[3]林子雨,杨冬青,王腾蛟. 用基于移动均值的索引实现时间序列相似查询[J]. 软件学报, 2008, 19(9): 2349-2361.Lin Zi-yu, Yang Dong-qing, and Wang Teng-jiao. Similarity search of time series with moving average based indexing[J].Journal of Software.2008, 19(9):2349-2361[4]靳碧,荣冈. BT: 一种快速序列搜索算法[J]. 浙江大学学报(工学版), 2007, 41(4): 621-625.Jin Bi and Rong Gang. BT: fast sequence search algorithm[J]. Journal of Zhejiang University (Engineering Science), 2007, 41(4): 621-625.[5]Johannes Afalg.[J].Hans-Peter Kriegel, and Peer Krger, et al.. Probabilistic similarity search for uncertain time series[C]. Proceedings of the 21st International Conference on Scientific and Statistical Database Management (SSDBM), New Orleans, LA, USA, Jun. 2-.2009,:-[6]Al-Naymat G and Taheri J. Effects of dimensionality reduction techniques on time series similarity measurement[C]. The 6th ACS/IEEE International Conference on Computer Systems and Applications, Doha, Qatar, Mar.31-Apr.4, 2008: 188-195.[7]Shatkay H and Zdonik S B. Approximate queries and representations for large data sequences[C]. Proceedings of the 12th International Conference on Data Engineering, New Orleans, Louisiana, Feb.26-Mar.1, 1996: 536-545.[8]王达,荣冈. 时间序列的模式距离[J]. 浙江大学学报(工学版), 2004, 38(7): 795-798.Wang Da and Rong Gang. Pattern distance of time series[J]. Journal of Zhejiang University (Engineering Science), 2004, 38(7): 795-798.[9]董晓莉,顾成奎,王正欧. 基于形态的时间序列相似性度量研究[J].电子与信息学报.2007, 29(5):1228-1231浏览Dong Xiao-li, Gu Cheng-kui, and Wang Zheng-ou. Research on shape-based time series similarity measure[J].Journal of Electronics Information Technology.2007, 29(5):1228-1231[10]张鹏,李学仁,张建业,等. 时间序列的夹角距离及相似性搜索[J]. 模式识别与人工智能, 2008, 21(6): 763-767.Zhang Peng, Li Xue-ren, and Zhang Jian-ye, et al.. Included angle distance of time series and similarity search[J]. Pattern Recognition and Artificial Intelligence, 2008, 21(6): 763-767.[11]Keogh E J. Fast similarity search in the presence of longitudinal scaling in time series databases[C]. Proceedings of the 9th International Conference on Tools with Artificial Intelligence, Newport Beach, CA, USA, Nov.3-8, 1997: 578-584.[12]Keogh E J, Chu S, and Hart D, et al.. Segmentation Time Series: A Survey and Novel Approach[M]. Data Mining in Time Series Databases. Singapore: World Scientific Publishing Co., 2004: 1-22.[13]Keogh E J and Pazzani M J. Relevance feedback retrieval of time series data[C]. Proceedings of the 22nd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, Berkeley, CA, USA, Aug. 15-19, 1999: 183-190.[14]Chang Pei-chann, Fan Chin-yuan, and Liu Chen-hao. Integrating a piecewise linear representation method and a neural network model for stock trading points prediction[J]. IEEE Transactions on System, Man, and Cybernetics, 2009, 30(1): 80-92.[15]Perng C S, Wang H, and Zhang S R. Landmarks: a new model for similarity-based pattern querying in time series databases[C]. Proceedings of the 16th International Conference on Data Engineering, San Diego, CA, USA, Feb.28-Mar.3, 2000: 33-42.[16]Phetking C, Sap M, and Selamat A. Identifying zigzag based perceptually important points for indexing financial time series[C]. Proceedings of the 8th International Conference on Cognitive Informatics, Hong Kong, China, Jun. 15-17, 2009: 295-301.[17]Kirkpatrick C D and Dahlquist J R. Technical Analysis: The Complete Resource for Financial Market Technicians[M]. 1st edition, Canada: Financial Time Prentice Hall, 2006: 11-12.[18]Pratt K and Fink E. Search for patterns in compressed time series[J]. International Journal of Image and Graphics, World Scientific, 2000, 2(1): 89-106.

施引文献

资源附件(0)

访问统计