Skip to main content

Classifying Motion Time Series Using Neural Networks

  • Conference paper
Advances in Multimedia Information Processing - PCM 2006 (PCM 2006)

Part of the book series: Lecture Notes in Computer Science ((LNISA,volume 4261))

Included in the following conference series:

  • 780 Accesses

Abstract

This paper proposes an effective time-series classification model based on the Neural Networks. Classification under this model consists of three phases, namely data preprocessing, training, and testing. The main contributions of the paper are described as following: We propose a feature extraction algorithm, which involves computation of finite difference of sequences, for preprocessing. We employ two different types of Neural Networks for training and testing. The results of the experiments on real univariate motion capture data and synthetic data show that our approach is effective in providing good performance in terms of accuracy. It is therefore a promising method for classifying time-series, in particular for univariate human motion capture data.

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

Access this chapter

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Similar content being viewed by others

References

  1. Berndt, D.J., Clifford, J.: Using dynamic time warping to find patterns in time series. In: KDD Workshop, pp. 359–370 (1994)

    Google Scholar 

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

    Google Scholar 

  3. Geurts, P.: Pattern extraction for time series classification. In: Siebes, A., De Raedt, L. (eds.) PKDD 2001. LNCS (LNAI), vol. 2168, pp. 115–127. Springer, Heidelberg (2001)

    Chapter  Google Scholar 

  4. Kalpakis, K., Gada, D., Puttagunta, V.: Distance measures for effective clustering of arima time-series. In: ICDM, pp. 273–280 (2001)

    Google Scholar 

  5. Kehagias, A., Petridis, V.: Predictive modular neural networks for time series classification. Neural Networks 10, 31–49 (1997)

    Article  Google Scholar 

  6. Keogh, E., Pazzani, M.: An enhanced representation of time series which allows fast and accurate classification, clustering and relevance feedback. In: KDD, pp. 239–243 (1998)

    Google Scholar 

  7. Keogh, E.J., Palpanas, T., Zordan, V., Gunopulos, D., Cardle, M.: Indexing large human-motion databases. In: VLDB, pp. 780–791 (2004)

    Google Scholar 

  8. Keogh, E.J., Smyth, P.: A probabilistic approach to fast pattern matching in time series databases. In: KDD, pp. 24–30 (1997)

    Google Scholar 

  9. Pao, Y.-H., Takefuji, Y.: Functional-link net computing: theory, system architecture, and functionalities. IEEE Computer 25(5), 76–79 (1992)

    Google Scholar 

  10. Ratanamahatana, C.A., Keogh, E.: The gun-point dataset. UCR Time Series Data, http://www.cs.ucr.edu/~eamonn/time_series_data/

  11. Theodoridis, S., Koutroumbas, K.: Pattern Recognition, 2nd edn. Elsevier Academic Press, San Diego (2003)

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Editor information

Editors and Affiliations

Rights and permissions

Reprints and permissions

Copyright information

© 2006 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Shou, L., Gao, G., Chen, G., Dong, J. (2006). Classifying Motion Time Series Using Neural Networks. In: Zhuang, Y., Yang, SQ., Rui, Y., He, Q. (eds) Advances in Multimedia Information Processing - PCM 2006. PCM 2006. Lecture Notes in Computer Science, vol 4261. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11922162_70

Download citation

  • DOI: https://doi.org/10.1007/11922162_70

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-48766-1

  • Online ISBN: 978-3-540-48769-2

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics