Learning DTW-Preserving Shapelets

  • Arnaud Lods
  • Simon MalinowskiEmail author
  • Romain Tavenard
  • Laurent Amsaleg
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10584)


Dynamic Time Warping (DTW) is one of the best similarity measures for time series, and it has extensively been used in retrieval, classification or mining applications. It is a costly measure, and applying it to numerous and/or very long times series is difficult in practice. Recently, Shapelet Transform (ST) proved to enable accurate supervised classification of time series. ST learns small subsequences that well discriminate classes, and transforms the time series into vectors lying in a metric space. In this paper, we adopt the ST framework in a novel way: we focus on learning, without class label information, shapelets such that Euclidean distances in the ST-space approximate well the true DTW. Our approach leads to an ubiquitous representation of time series in a metric space, where any machine learning method (supervised or unsupervised) and indexing system can operate efficiently.


  1. 1.
    Bagnall, A., Lines, J., Vickers, W., Keogh, E.: The UEA and UCR time series classification repository.
  2. 2.
    Bromley, J., Guyon, I., LeCun, Y., Säckinger, E., Shah, R.: Signature verification using a “siamese” time delay neural network. In: Advances in Neural Information Processing Systems, pp. 737–744 (1994)Google Scholar
  3. 3.
    Duchi, J., Hazan, E., Singer, Y.: Adaptive subgradient methods for online learning and stochastic optimization. JMLR 12, 2121–2159 (2011)MathSciNetzbMATHGoogle Scholar
  4. 4.
    Esling, P., Agon, C.: Time-series data mining. ACM Comput. Surv. 45(1) (2012)Google Scholar
  5. 5.
    Grabocka, J., Schilling, N., Wistuba, M., Schmidt-Thieme, L.: Learning time-series shapelets. In: Proceedings of KDD (2014)Google Scholar
  6. 6.
    Hills, J., Lines, J., Baranauskas, E., Mapp, J., Bagnall, A.: Classification of time series by shapelet transformation. DMKD 28(4), 851–881 (2014)MathSciNetzbMATHGoogle Scholar
  7. 7.
    Keogh, E., Ratanamahatana, C.A.: Exact indexing of dynamic time warping. KAIS 7, 358–386 (2005)Google Scholar
  8. 8.
    Lei, Q., Yi, J., Vaculín, R., Wu, L., Dhillon, I.S.: Similarity preserving representation learning for time series analysis (2017).
  9. 9.
    Lemire, D.: Faster retrieval with a two-pass dynamic-time-warping lower bound. Pattern Recogn. 42(9), 2169–2180 (2009)CrossRefzbMATHGoogle Scholar
  10. 10.
    Paparrizos, J., Gravano, L.: k-Shape: efficient and accurate clustering of time series. In: Proceedings of SIGMOD (2015)Google Scholar
  11. 11.
    Rakthanmanon, T., Campana, B., Mueen, A., Batista, G., Westover, B., Zhu, Q., Zakaria, J., Keogh, E.: Searching and mining trillions of time series subsequences under dynamic time warping. In: Proceedings of KDD (2012)Google Scholar
  12. 12.
    Tan, C.W., Webb, G.I., Petitjean, F.: Indexing and classifying gigabytes of time series under time warping. In: Proceedings of SIAM ICDM (2017)Google Scholar
  13. 13.
    Ye, L., Keogh, E.: Time series shapelets: a new primitive for data mining. In: Proceedings of KDD (2009)Google Scholar
  14. 14.
    Zakaria, J., Mueen, A., Keogh, E.: Clustering time series using unsupervised-shapelets. In: Proceedings of ICDM (2012)Google Scholar
  15. 15.
    Zhang, Q., Wu, J., Yang, H., Tian, Y., Zhang, C.: Unsupervised feature learning from time series. In: Proceedings of IJCAI (2016)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  • Arnaud Lods
    • 1
  • Simon Malinowski
    • 2
    Email author
  • Romain Tavenard
    • 3
  • Laurent Amsaleg
    • 4
  1. 1.IRISARennesFrance
  2. 2.Univ. Rennes 1, IRISARennesFrance
  3. 3.Univ. Rennes 2, CNRS, UMR LETG, IRISARennesFrance
  4. 4.CNRS-IRISARennesFrance

Personalised recommendations