Data Mining and Knowledge Discovery

, Volume 22, Issue 1–2, pp 149–182 | Cite as

Time series shapelets: a novel technique that allows accurate, interpretable and fast classification

Open Access


Classification of time series has been attracting great interest over the past decade. While dozens of techniques have been introduced, recent empirical evidence has strongly suggested that the simple nearest neighbor algorithm is very difficult to beat for most time series problems, especially for large-scale datasets. While this may be considered good news, given the simplicity of implementing the nearest neighbor algorithm, there are some negative consequences of this. First, the nearest neighbor algorithm requires storing and searching the entire dataset, resulting in a high time and space complexity that limits its applicability, especially on resource-limited sensors. Second, beyond mere classification accuracy, we often wish to gain some insight into the data and to make the classification result more explainable, which global characteristics of the nearest neighbor cannot provide. In this work we introduce a new time series primitive, time series shapelets, which addresses these limitations. Informally, shapelets are time series subsequences which are in some sense maximally representative of a class. We can use the distance to the shapelet, rather than the distance to the nearest neighbor to classify objects. As we shall show with extensive empirical evaluations in diverse domains, classification algorithms based on the time series shapelet primitives can be interpretable, more accurate, and significantly faster than state-of-the-art classifiers.


Time series Data mining Classification Decision tree 


  1. Anon (1525) Founders’ and benefectors’ book of Tewkesbury Abbey, in Latin England.
  2. Berndt DJ, Clifford J (1994) Using dynamic time warping to find patterns in time series. AAAI-94 workshop on knowledge discovery in databases, Seattle, Washington, 31 July 1994Google Scholar
  3. Breiman L, Friedman J, Olshen RA, Stone CJ (1984) Classification and regression trees. Wadsworth, Belmont, CAMATHGoogle Scholar
  4. Briandet R, Kemsley EK, Wilson RH (1996) Discrimination of arabica and robusta in instant coffee by fourier transform infrared spectroscopy and chemometrics. Food Chem 44(1): 170–174CrossRefGoogle Scholar
  5. Chiu B, Keogh E, Lonardi S (2003) Probabilistic discovery of time series motifs. In: Proceedings of the ninth ACM SIGKDD international conference on knowledge discovery and data mining, Washington, DC, 24–27 Aug 2003. KDD ‘03, ACM, New York, NY, pp 493–498Google Scholar
  6. CMU Graphics Lab Motion Capture Database
  7. 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. In: Proceedings of the VLDB endowment, Aug 2008, vol 1, 2. pp 1542–1552Google Scholar
  8. Faloutsos C, Ranganathan M, Manolopoulos Y (1994) Fast subsequence matching in time-series databases. ACM SIGMOD Record 23, June 1994, vol 2. pp 419–429Google Scholar
  9. Geurts P (2001) Pattern extraction for time series classification. In: Raedt LD, Siebes A (eds) Proceedings of the 5th European conference on principles of data mining and knowledge discovery, Sept 03–05, 2001 (Lecture notes in computer science), vol 2168. Springer-Verlag, London, pp 115– 127Google Scholar
  10. Gramm J, Guo J, Niedermeier R (2003) On exact and approximation algorithms for distinguishing substring selection. In: Proceedings of 14th fundamentals of computation theory (Lecture notes in computer science), vol 2751. Springer-Verlag, London, pp 963–971Google Scholar
  11. Jeong MK, Lu JC, Huo X, Vidakovic B, Chen D (2006) Wavelet-based data reduction techniques for process fault detection. Technometrics 48(1): 26–40CrossRefMathSciNetGoogle Scholar
  12. Kadous MW (1999) Learning comprehensible descriptions of multivariate time series. In: Bratko I, Dzeroski S (eds) Proceedings of the sixteenth international conference on machine learning, June 27–30, 1999. Morgan Kaufmann Publishers, San Francisco, CA, pp 454–463Google Scholar
  13. Keogh E, Kasetty S (2002) On the need for time series data mining benchmarks: a survey and empirical demonstration. In: Proceedings of the eighth ACM SIGKDD international conference on knowledge discovery and data mining, Edmonton, Alberta, Canada, 23–26 July 2002. KDD ‘02. ACM, New York, NY, pp 102–111Google Scholar
  14. Keogh E, Ratanamahatana CA (2005) Exact indexing of dynamic time warping knowledge. Inf Syst 7(3): 358–386Google Scholar
  15. Keogh E, Wei L, Xi X, Lee S, Vlachos M (2006) LB_Keogh supports exact indexing of shapes under rotation invariance with arbitrary representations and distance measures. In: Dayal U, Whang K, Lomet D, Alonso G, Lohman G, Kersten M, Cha SK, Kim Y (eds) Proceedings of the 32nd international conference on very large data bases, Seoul, Korea, 12–15 Sept 2006. Very large data bases. VLDB Endowment, pp 882–893Google Scholar
  16. Koschorreck W, Werner W (eds) (1981) Facsimile edition with commentary: Kommentar zum faksimile des codex manesse: Die grosse Heidelberger LiederhandschriftGoogle Scholar
  17. Lang W, Morse M, Patel JM (2009) Dictionary-based compression for long time-series similarity. IEEE transactions on knowledge and data engineering, 15 Oct 2009Google Scholar
  18. Lin J, Keogh E, Wei L, Lonardi S (2007) Experiencing SAX: a novel symbolic representation of time series. Data Min Knowl Discov 15(2): 107–144CrossRefMathSciNetGoogle Scholar
  19. Martinez AM, Kak AC (2001) PCA versus LDA. IEEE Trans Patt Anal Mach Intel 23(2): 228–233CrossRefGoogle Scholar
  20. Montagu JA (1840) A guide to the study of heraldry. W. Pickering, London.
  21. Rodríguez JJ, Alonso CJ (2004) Interval and dynamic time warping-based decision trees. In: Proceedings of the 2004 ACM symposium on applied computing, Nicosia, Cyprus, March 14–17, 2004. SAC ‘04. ACM, New York, NY, pp 548–552Google Scholar
  22. Roverso D (2000) Multivariate temporal classification by windowed wavelet decomposition and recurrent neural networks. In: 3rd ANS international topical meeting on nuclear plant instrumentation, control and human-machine interface, 2000Google Scholar
  23. Salzberg SL (1997) On comparing classifiers: pitfalls to avoid and a recommended approach. Data Min Knowl Discov 1: 317–327CrossRefGoogle Scholar
  24. Veloso A, Meira W, Zaki MJ (2006) Lazy associative classification. In: Proceedings of the sixth international conference on Data mining, Dec 18–22, 2006. ICDM. IEEE Computer Society, Washington, DC, pp 645–654Google Scholar
  25. Wikipedia description of coffee:
  26. Wilcoxon F (1945) Individual comparisons by ranking methods. Biometrics 1: 80–83CrossRefGoogle Scholar
  27. Wilson RH, Goodfellow BG (1994) Mid-infrared spectroscopy. Spectroscopic Techniques for Food AnalysisGoogle Scholar
  28. Xi X, Keogh E, Shelton C, Wei L, Ratanamahatana CA (2006) Fast time series classification using numerosity reduction. In: Proceedings of the 23rd international conference on machine learning, Pittsburgh, Pennsylvania, June 25–29, 2006. ICML ‘06, vol 148. ACM, New York, NY, pp 1033– 1040Google Scholar
  29. Yamada Y, Suzuki E, Yokoi H, Takabayashi K (2003) Decision-tree induction from time-series data based on a standard example split test. In: Proceedings of the 20th international conference on machine learning, pp 840–847Google Scholar
  30. Ye L (2009) The time series shapelet Webpage.
  31. 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, Paris, France, June 28 to July 01 2009. KDD ‘09. ACM, New York, NY, pp 947–956Google Scholar

Copyright information

© The Author(s) 2010

Authors and Affiliations

  1. 1.Department of Computer Science and EngineeringUniversity of CaliforniaRiversideUSA

Personalised recommendations