Feature representation of multivariate time series based on correlation among variables

被引：0
|
作者：
Li, Hai-Lin [1 ]
机构：
[1] School of Business Management, Huaqiao University, Quanzhou
来源：
Kongzhi yu Juece/Control and Decision | 2015年 / 30卷 / 03期
关键词：
Data mining; Feature representation; Multivariate time series; Principal component analysis;
D O I：
10.13195/j.kzyjc.2014.0132
中图分类号：
学科分类号：
摘要：
The property of high dimensionality impacts on the process and results in the field of time series data mining, and the traditional methods about principal component analysis have some limitations to represent multivariate time series. Therefore, a feature representation of multivariate time series based on correlation among variables is proposed. The distribution and relationships among variants of every time series are described by the covariance matrix, and principal components are extracted from an integrated covariance matrix by principal component analysis. In this way, the dimensionality of multivariate time series can be reduced and the features can be represented. The experimental results show that the proposed method not only improves the quality of multivariate time series data mining but also efficiently mines on the data with different lengths. ©, 2015, Northeast University. All right reserved.
引用
收藏
页码：441 / 447
页数：6
相关论文
共 15 条
  • [1] Esling P., Agon C., Time-series data mining, ACM Computing Surverys, 45, 1, pp. 11-12, (2012)
  • [2] Wang Z., Chan L., Learning causal relations in multivariate time series data, ACM Trans on Int Systems and Technology, 3, 4, pp. 71-76, (2012)
  • [3] Li H., Yang L., Guo C., Improved piecewise vector quantized approximation based on normalized time subsequences, Measurement, 46, 9, pp. 3429-3439, (2013)
  • [4] Yan Q.Y., Xia S.X., An piecewise linear fitting algorithm for infinite time series, Acta Electronica Sinica, 28, 2, pp. 443-448, (2010)
  • [5] Li H.L., Yang L.B., Method of dimensionality reduction and feature representation for time series, Control and Decision, 28, 11, pp. 1718-1722, (2013)
  • [6] Yang K., Shahabi C., An efficient k nearest neighbor search for multivariate time series, Information and Computation, 205, 1, pp. 65-98, (2007)
  • [7] Weng X., Shen J., Classification of multivariate time series using two-dimensional singular value decomposition, Knowledge-Based Systems, 21, 7, pp. 535-539, (2008)
  • [8] Han M., Li D.C., Multiple time series correlation extraction and prediction based on EOF-SVD model, J of System Simulation, 7, pp. 1669-1672, (2008)
  • [9] Baragona R., Battaglia F., Outliers detection in multivariate time series by independent component analysis, Neural Computation Archive, 19, 7, pp. 1962-1984, (2007)
  • [10] Li Z.X., Guo J.S., Hui X.B., Et al., Dimension reduction method for multivariate time series based on common principal component, Control and Decision, 28, 4, pp. 531-536, (2013)
12