Advertisement

Scale Invariant Multi-length Motif Discovery

  • Yasser Mohammad
  • Toyoaki Nishida
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8482)

Abstract

Discovering approximately recurrent motifs (ARMs) in timeseries is an active area of research in data mining. Exact motif discovery was later defined as the problem of efficiently finding the most similar pairs of timeseries subsequences and can be used as a basis for discovering ARMs. The most efficient algorithm for solving this problem is the MK algorithm which was designed to find a single pair of timeseries subsequences with maximum similarity at a known length. Available exact solutions to the problem of finding top K similar subsequence pairs at multiple lengths (which can be the basis of ARM discovery) are not scale invariant. This paper proposes a new algorithm for solving this problem efficiently using scale invariant distance functions and applies it to both real and synthetic dataset.

Keywords

Distance Calculation Motif Discovery Triangular Inequality Motif Length Scanning Phase 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Chiu, B., Keogh, E., Lonardi, S.: Probabilistic discovery of time series motifs. In: ACM KDD 2003, pp. 493–498 (2003)Google Scholar
  2. 2.
    CMU: Cmu motion capture dataset, http://mocap.cs.cmu.edu
  3. 3.
    Keogh, E., Lin, J., Fu, A.: Hot sax: Efficiently finding the most unusual time series subsequence. In: IEEE ICDM, pp. 8–16 (November 2005)Google Scholar
  4. 4.
    Keogh, E., Kasetty, S.: On the need for time series data mining benchmarks: A survey and empirical demonstration. Data Mining and Knowledge Discovery 7(4), 349–371 (2003)Google Scholar
  5. 5.
    Minnen, D., Starner, T., Essa, I., Isbell, C.: Improving activity discovery with automatic neighborhood estimation. In: IJCAI (2007)Google Scholar
  6. 6.
    Mohammad, Y., Nishida, T.: Exact discovery of length-range motifs. In: Nguyen, N.T., Attachoo, B., Trawiński, B., Somboonviwat, K. (eds.) ACIIDS 2014, Part II. LNCS, vol. 8398, pp. 23–32. Springer, Heidelberg (2014)Google Scholar
  7. 7.
    Mohammad, Y., Nishida, T.: Constrained motif discovery in time series. New Generation Computing 27(4), 319–346 (2009)Google Scholar
  8. 8.
    Mohammad, Y., Nishida, T.: Fluid imitation: Discovering what to imitate. International Journal of Social Robotics 4(4), 369–382 (2012)Google Scholar
  9. 9.
    Mohammad, Y., Nishida, T.: Unsupervised discovery of basic human actions from activity recording datasets. In: Proceedings of the IEEE/SICE SII (2012)Google Scholar
  10. 10.
    Mohammad, Y., Nishida, T., Okada, S.: Unsupervised simultaneous learning of gestures, actions and their associations for human-robot interaction. In: Proceedings of IROS 2009, pp. 2537–2544 (2009)Google Scholar
  11. 11.
    Mueen, A.: Enumeration of time series motifs of all lengths. In: 2013 IEEE 13th International Conference on Data Mining (ICDM). IEEE (2013)Google Scholar
  12. 12.
    Mueen, A., Keogh, E.J., Zhu, Q., Cash, S., Westover, M.B.: Exact discovery of time series motifs. In: SDM, pp. 473–484 (2009)Google Scholar
  13. 13.
    Patel, P., Keogh, E., Lin, J., Lonardi, S.: Mining motifs in massive time series databases. In: IEEE International Conference on Data Mining, pp. 370–377 (2002)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Yasser Mohammad
    • 1
    • 2
  • Toyoaki Nishida
    • 2
  1. 1.Assiut UniversityEgypt
  2. 2.Kyoto UniversityJapan

Personalised recommendations