Skip to main content
SpringerLink
Log in

Sequential data classification by dynamic state warping

  • Regular Paper
  • Published:
Knowledge and Information Systems Aims and scope Submit manuscript

Abstract

The ubiquity of sequences in many domains enhances significant recent interest in sequence learning, for which a basic problem is how to measure the distance between sequences. Dynamic time warping (DTW) aligns two sequences by nonlinear local warping and returns a distance value. DTW shows superior ability in many applications, e.g. video, image, etc. However, in DTW, two points are paired essentially based on point-to-point comparisons without considering the autocorrelation of sequences. Thus, points with different semantic meanings, e.g. peaks and valleys, may be matched providing their coordinate values are similar. As a result, DTW may be sensitive to noise and poorly interpretable. This paper proposes an improved alignment method, dynamic state warping (DSW). DSW integrates the dynamic information of sequences into DTW by converting each time point into a latent state. Alignment is performed by using the state sequences. Thus, DSW is able to yield alignment that is semantically more interpretable than that of DTW. Using one nearest neighbour classifier, DSW shows significant improvement on classification accuracy in comparison with Euclidean distance (68/85 wins), DTW (70/85 wins) and its variants. We also empirically demonstrate that DSW is more robust and scales better to long sequences than Euclidean distance and DTW.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16
Fig. 17
Fig. 18
Fig. 19

Similar content being viewed by others

Notes

  1. Since our concern is the effectiveness of incorporating the dynamics into time point representation instead of the parameter tuning procedure, this motivates us to compare DSW and DTW both without constraining the size of the warping window parameter.

  2. By employing constraints on the alignment, the complexity of DTW alignment could be linear [11]. In this paper, we focus on the effectiveness of distance measurement based on states of time points, instead of the original time points. To ease the comparison, we will not consider constraint strategy. But it is straightforward to incorporate alignment constraints in DSW.

References

  1. Batal I, Cooper GF, Fradkin D, Harrison J, Moerchen F, Hauskrecht M (2016) An efficient pattern mining approach for event detection in multivariate temporal data. Knowl Inf Syst 46(1):115–150

    Article  Google Scholar 

  2. Jo Y, Loghmanpour N, Rosé CP (2015) Time series analysis of nursing notes for mortality prediction via a state transition topic model. In: Proceedings of the 24th ACM international on conference on information and knowledge management, ACM, pp 1171–1180

  3. Chen H, Tino P, Rodan A, Yao X (2014) Learning in the model space for cognitive fault diagnosis. IEEE Trans Neural Netw Learn Syst 25(1):124–136

    Article  Google Scholar 

  4. Goroshin R, Bruna J, Tompson J, Eigen D, LeCun Y (2015) Unsupervised learning of spatiotemporally coherent metrics. In: Proceedings of the IEEE international conference on computer vision, pp 4086–4093

  5. Pei W, Tax DM, van der Maaten L (2016) Modeling time series similarity with siamese recurrent networks. arXiv preprint arXiv:1603.04713

  6. Chen H, Tang F, Tino P, Yao X (2013) Model-based kernel for efficient time series analysis. In: Proceedings of the 19th ACM SIGKDD international conference on knowledge discovery and data mining, ACM, pp 392–400

  7. Chen H, Tang F, Tino P, Cohn AG, Yao X (2015) Model metric co-learning for time series classification. In: Proceedings of the twenty-fourth international joint conference on artificial intelligence, AAAI Press, pp 3387–3394

  8. Aminikhanghahi S, Cook DJ (2017) A survey of methods for time series change point detection. Knowl Inf Syst 51(2):339–367

    Article  Google Scholar 

  9. Bagnall A, Bostrom A, Large J, Lines J (2016) The great time series classification bake off: an experimental evaluation of recently proposed algorithms. Extended version. arXiv preprint arXiv:1602.01711

  10. Ye L, Keogh E (2009) Time series shapelets: a new primitive for data mining. In: Proceedings of the 15th ACM SIGKDD international conference on knowledge discovery and data mining, ACM, pp 947–956

  11. Rakthanmanon T, Campana B, Mueen A, Batista G, Westover B, Zhu Q, Zakaria J, Keogh E (2012) Searching and mining trillions of time series subsequences under dynamic time warping. In: Proceedings of the 18th ACM SIGKDD international conference on knowledge discovery and data mining, ACM, pp 262–270

  12. Batista GE, Keogh EJ, Tataw OM, de Souza VM (2014) CID: an efficient complexity-invariant distance for time series. Data Min Knowl Disc 28(3):634–669

    Article  MathSciNet  MATH  Google Scholar 

  13. Berndt DJ, Clifford J (1994) Using dynamic time warping to find patterns in time series. In: AAAI workshop on KDD, Seattle, WA, vol 10, pp 359–370

  14. Keogh EJ, Pazzani MJ (2001) Derivative dynamic time warping. In: SDM, SIAM, vol 1, pp 5–7

  15. Zhou F, De la Torre F (2016) Generalized canonical time warping. IEEE Trans Pattern Anal Mach Intell 38(2):279–294

    Article  Google Scholar 

  16. Kogan JA, Margoliash D (1998) Automated recognition of bird song elements from continuous recordings using dynamic time warping and hidden markov models: A comparative study. J Acoust Soc Am 103(4):2185–2196

    Article  Google Scholar 

  17. Ding H, Trajcevski G, Scheuermann P, Wang X, Keogh E (2008) Querying and mining of time series data: experimental comparison of representations and distance measures. Proc VLDB Endow 1(2):1542–1552

    Article  Google Scholar 

  18. Begum N, Ulanova L, Wang J, Keogh E (2015) Accelerating dynamic time warping clustering with a novel admissible pruning strategy. In: Proceedings of the 21th ACM SIGKDD international conference on knowledge discovery and data mining, ACM, pp 49–58

  19. Shariat S, Pavlovic V (2016) Robust time-series retrieval using probabilistic adaptive segmental alignment. Knowl Inf Syst 49(1):91–119

    Article  Google Scholar 

  20. Ye L, Keogh E (2011) Time series shapelets: a novel technique that allows accurate, interpretable and fast classification. Data Min Knowl Disc 22(1–2):149–182

    Article  MathSciNet  MATH  Google Scholar 

  21. Chen Y, Keogh E, Hu B, Begum N, Bagnall A, Mueen A, Batista G (2015) The UCR time series classification archive. www.cs.ucr.edu/~eamonn/time_series_data/

  22. Jeong Y-S, Jeong MK, Omitaomu OA (2011) Weighted dynamic time warping for time series classification. Pattern Recogn 44(9):2231–2240

    Article  Google Scholar 

  23. Faloutsos C, Ranganathan M, Manolopoulos Y (1994) Fast subsequence matching in time-series databases. In: ACM SIGMOD Record, ACM, pp 419–429

  24. Lemire D (2009) Faster retrieval with a two-pass dynamic-time-warping lower bound. Pattern Recogn 42(9):2169–2180

    Article  MATH  Google Scholar 

  25. Garreau D, Lajugie R, Arlot S, Bach F (2014) Metric learning for temporal sequence alignment. In: Advances in neural information processing systems, pp 1817–1825

  26. Petitjean F, Forestier G, Webb GI, Nicholson AE, Chen Y, Keogh E (2014) Dynamic time warping averaging of time series allows faster and more accurate classification. In: 2014 IEEE international conference on data mining, IEEE, pp 470–479

  27. Neubrandt D, Buza K (2017) Projection-based person identification. In: International conference on computer recognition systems, Springer, Berlin, pp 221–228

  28. Meszlényi RJ, Hermann P, Buza K, Gál V, Vidnyánszky Z (2017) Resting state fMRI functional connectivity analysis using dynamic time warping. Front Neurosci 11:75

    Article  Google Scholar 

  29. Mikolov T, Sutskever I, Chen K, Corrado GS, Dean J (2013) Distributed representations of words and phrases and their compositionality. In: Advances in neural information processing systems, pp 3111–3119

  30. Jaeger H (2002) Tutorial on training recurrent neural networks, covering BPPT, RTRL, EKF and the “echo state network” approach, vol 5. GMD-Forschungszentrum Informationstechnik, Bonn

    Google Scholar 

  31. Jaeger H (2001) Short term memory in echo state networks. Bonn, GMD-Forschungszentrum Informationstechnik

    Google Scholar 

  32. Natschläger T, Maass W, Markram H (2002) The “liquid computer”: a novel strategy for real-time computing on time series. In: Special issue on foundations of information processing of TELEMATIK, vol 8, no LNMC-ARTICLE-2002-005, pp 39–43

  33. LukošEvičIus M, Jaeger H (2009) Reservoir computing approaches to recurrent neural network training. Comput Sci Rev 3(3):127–149

    Article  MATH  Google Scholar 

  34. Maaten L (2011) Learning discriminative fisher kernels. In: Proceedings of the 28th international conference on machine learning, pp 217–224

  35. Srivastava N, Mansimov E, Salakhudinov R (2015) Unsupervised learning of video representations using lstms. In: International conference on machine learning, pp 843–852

  36. Rodan A, Tiňo P (2012) Simple deterministically constructed cycle reservoirs with regular jumps. Neural Comput 24(7):1822–1852

    Article  MathSciNet  Google Scholar 

  37. Höppner F (2017) Improving time series similarity measures by integrating preprocessing steps. Data Min Knowl Disc 31(3):851–878

    Article  MathSciNet  Google Scholar 

  38. Grewal MS (2011) Kalman filtering. In: Lovric M (ed) International encyclopedia of statistical science. Springer, Berlin, pp 705–708

  39. Anissa S, Hassene S, Zouhair M (2013) Efficient speech denoising applied to colored noise based dynamic low-pass filter supervised by cascade neural networks. In: 2013 International conference on electrical engineering and software applications, IEEE, pp 1–5

  40. Jaeger H (2001) The echo state approach to analysing and training recurrent neural networks-with an erratum note. Bonn, Germany: German National Research Center for Information Technology GMD Technical Report vol 148, p 34

  41. Van der Maaten L, Hinton G (2008) Visualizing data using t-SNE. J Mach Learn Res 9(2579–2605):85

    MATH  Google Scholar 

Download references

Acknowledgements

We would like to thank the anonymous reviewers for their constructive comments and suggestions to improve the quality of the paper. This work is supported in part by the National Key Research and Development Program of China (Grant No. 2016YFB1000905), the National Natural Science Foundation of China (Grant Nos. 91546116 and 61673363) and USTC Young Innovation Grant (WK2150110001).

Author information

Authors and Affiliations

  1. School of Computer Science and Technology, University of Science and Technology of China, Hefei, China

    Zhichen Gong & Huanhuan Chen

Authors
  1. Zhichen Gong
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Huanhuan Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Huanhuan Chen.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Gong, Z., Chen, H. Sequential data classification by dynamic state warping. Knowl Inf Syst 57, 545–570 (2018). https://doi.org/10.1007/s10115-017-1139-9

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10115-017-1139-9

Keywords

Search

Navigation