基于加权动态时间弯曲的多元时间序列相似性匹配方法<sup>*</sup>

doi:10.16451/j.cnki.issn1003-6059.201704003

摘要
图/表
参考文献
相关文章 (2)

全文: PDF (4802 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要针对常用方法忽略变量相关性和局部形状特性问题,提出基于加权动态时间弯曲的多元时间序列相似性匹配方法(CPCA-SWDTW).首先,在原加权动态时间弯曲算法基础上,引入形态因子,提出基于形态特征的加权动态时间弯曲算法(SWDTW).然后,提取多元时间序列的主成分作为模式表示,消除变量间的相关性,同时将方差贡献率作为相应主成分的权重.在此基础上,运用SWDTW,度量多元时间序列间的相似度.最后,通过相似性搜索实验表明,CPCA-SWDTW具有较好的准确性和鲁棒性.敏感性分析说明CPCA-SWDTW在一定程度上受到权重函数参数的影响.

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	叶燕清
	杨克巍
	姜江
	葛冰峰
	豆亚杰

关键词 ：多元时间序列, 相似性匹配, 共同主成分分析, 加权动态时间弯曲

Abstract：In most of the current methods, the close correlation between variables and the shape characteristics of time series is neglected. In this paper, a similarity matching method for multivariate time series is proposed based on combined principal component analysis method and a shape-based improved weighted dynamic time warping algorithm(CPCA-SWDTW). Firstly, a shape coefficient is introduced and a shape based weighted dynamic time warping(SWDTW) algorithm is presented. Next, the principal components of multivariate time series are extracted as the representation, and thus the variable correlations can be eliminated. Besides, the variance devoting rate of each principal component is considered as the weight of each series. On the basis of the proposed representation, SWDTW is used to measure the similarity between transformed multivariate time series. Finally, the results of similarity search experiment show that CPCA-SWDTW is more efficient and robust. Moreover, the parameter sensitivity analysis experiment show that CPCA-SWDTW can be affected by the parameters in weight function to some extent.

Key words： Multivariate Time Series Similarity Matching Combined Principal Component Analysis Weighted Dynamic Time Warping

收稿日期: 2016-10-15

ZTFLH:

TP 391

基金资助:国家自然科学基金项目(No.71671186,71571185,71501182)资助

作者简介: 叶燕清,女,1990年生,博士研究生,主要研究方向为模式识别、数据挖掘、知识图谱.E-mail:yeyanqing@nudt.edu.cn.
杨克巍,男,1977年生,博士,教授,主要研究方向为国际采办、需求分析与建模、智能代理仿真.E-mail:kayyang27@163.com.
姜江(通讯作者),男,1981年生,博士,副教授,主要研究方向为证据推理、不确定性决策、风险分析.E-mail:jiangjiangnudt@163.com.
葛冰峰,男,1983年生,博士,讲师,主要研究方向为冲突消解、多目标规划.E-mail:bingfengge@163.com.
豆亚杰,男,1987年生,博士,讲师,主要研究方向为组合选择决策、多目标规划.E-mail:yajie_nudt@163.com.

引用本文:

叶燕清，杨克巍，姜江，葛冰峰，豆亚杰. 基于加权动态时间弯曲的多元时间序列相似性匹配方法^*[J]. 模式识别与人工智能, 2017, 30(4): 314-327. YE Yanqing, YANG Kewei, JIANG Jiang, GE Bingfeng, DOU Yajie. Multivariate Time Series Similarity Matching Method Based on Weighted Dynamic Time Warping Algorithm. , 2017, 30(4): 314-327.

链接本文:

http://manu46.magtech.com.cn/Jweb_prai/CN/10.16451/j.cnki.issn1003-6059.201704003 或 http://manu46.magtech.com.cn/Jweb_prai/CN/Y2017/V30/I4/314

[1] 吴绍春,吴耿锋,王炜,等.寻找地震相关地区的时间序列相似性匹配算法.软件学报, 2006, 17(2): 185-192.
(WU S C, WU G F, WANG W, et al. A Time-Sequence Similarity Matching Algorithm for Seismological Relevant Zones. Journal of Software, 2006, 17(2): 185-192.)
[2] 陈立新,张磊.一种面向金融数据的基于ARIMA的模糊时间序列预测模型.现代计算机, 2014(5): 3-8, 13.
(CHEN L X, ZHANG L. A Fuzzy Time Series Forecasting Model for Financial Data Based on ARIMA. Modern Computer, 2014(5): 3-8, 13.)
[3] 顾昕辰,万定生,樊龙.基于Hadoop的水文时间序列相似性研究与应用.计算机与数字工程, 2014, 42(11): 1-6.
(GU X C, WAN D S, FAN L. Research and Application of Hydrological Time Series Similarity Based on Hadoop. Computer and Di-
gital Engineering, 2014, 42(11): 1-6.)
[4] LEHNERTZ K. Non-linear Time Series Analysis of Intracranial EEG Recordings in Patients with Epilepsy-An Overview. International Journal of Psychophysiology, 1999, 34(1): 45-52.
[5] 何晓旭.时间序列数据挖掘若干关键问题研究.博士学位论文.合肥:中国科学技术大学, 2014.
(HE X X. Research on Key Issues in Time Series Data Mining. Ph.D Dissertation. Hefei, China: University of Science and Technology of China, 2014.)
[6] 吴虎胜,张凤鸣,张超,等.多元时间序列的相似性匹配.应用科
学学报, 2013, 31(6): 643-649.
(WU H S, ZHANG F M, ZHANG C, et al. Matching Similar Pa-
tterns for Multivariate Time Series. Journal of Applied Sciences-Electronics and Information Engineering, 2013, 31(6): 643-649.)
[7] 管河山,姜青山,王声瑞.基于点分布特征的多元时间序列模式匹配方法.软件学报, 2009, 20(1): 67-79.
(GUAN H S, JIANG Q S, WANG S R. Pattern Matching Method Based on Point Distribution for Multivariate Time Series. Journal of Software, 2009, 20(1): 67-79.)
[8] YANG K, SHAHABI C. A PCA-Based Similarity Measure for Multivariate Time Series. IEEE Transactions on Knowledge and Data Engineering, 2005, 17(9): 235-242.
[9] YI B K, FALOUTSOS C. Fast Time Sequence Indexing for Arbitrary Lp Norms // Proc of the 26th International Conference on Very Large Data Bases. San Francesco, USA: Morgan Kaufmann Publishers, 2000: 385-394.
[10] STEM H, SHMUELI M, BERMAN S. Most Discriminating Segment-Longest Common Subsequence(MDSLCS) Algorithm for Dynamic Hand Gesture Classification. Pattern Recognition Letters, 2013, 34(15): 1980-1989.
[11] KEOGH E, RATANAMAHATANA C A. Exact Indexing of Dynamic Time Warping. Knowledge and Information Systems, 2005, 7(3): 358-386.
[12] KIM S W, PARK S, CHU W W. An Index-Based Approach for Similarity Search Supporting Time Warping in Large Sequence Databases // Proc of the 17th International Conference on Data Engineering. Washington, USA: IEEE, 2001: 607-614.
[13] GRECKI T, UCZAK M. Non-isometric Transforms in Time Series Classification Using DTW. Knowledge-Based Systems, 2014, 61: 98-108.
[14] JEONG Y S, JEONG M K, OMITAOMU O A. Weighted Dynamic Time Warping for Time Series Classification. Pattern Recognition, 2011, 44(9): 2231-2240.
[15] GRECKI T, UCZAK M. Multivariate Time Series Classification with Parametric Derivative Dynamic Time Warping. Expert Systems with Applications, 2015, 42(5): 2305-2312.
[16] RODRIGUEZ J J, ALONSO C J, MAESTRO J A. Support Vector Machines of Interval-Based Features for Time Series Classification. Knowledge-Based Systems, 2005, 18(4/5): 171-178.