Concatenated Decision Paths Classification for Datasets with Small Number of Class Labels

Ivan Mitzev, Nicolas H. Younan

Abstract

In recent years, the amount of collected information has rapidly increased, that has led to an increasing interest to time series data mining and in particular to the classification of these data. Traditional methods for classification are based mostly on distance measures between the time series and 1-NN classification. Recent development of classification methods based on time series shapelets- propose using small sub-sections of the entire time series, which appears to be most representative for certain classes. In addition, the shapelets-based classification method produces higher accuracies on some datasets because the global features are more sensitive to noise than the local ones. Despite its advantages the shapelets methods has an apparent disadvantage- slow training time. Varieties of algorithms were proposed to tackle this problem, one of which is the concatenated decision paths (CDP) algorithm. This algorithm as initially proposed works only with datasets with a number of class indexes higher than five. In this paper, we investigate the possibility to use CDP for datasets with less than five classes. We also introduce improvements that shorten the overall training time of the CDP method.

References

  1. Ye, L., Keogh, E. (2009). Time Series Shapelets: A New Primitive for Data Mining. In: KDD'09, June 29-July 1, Paris, France, 2009.
  2. He, Q., Dong, Z., Zhuang, F., Shang, T., Shi, Z. (2012). Fast Time Series Classification Based on Infrequent Shapelets. In: ICMA'12, Beijing, China, 2012.
  3. Rakthanmanon, T., Keogh, E. (2013). Fast Shapelets: A Scalable Algorithm for Discovering Time Series Shapelets. In: SDM, SIAM'13, February 25- March 1, Boston, USA, 2013.
  4. Grabocka, J., Wistuba, M., Schmidt-Thieme, L. (2015). Scalable Discovery of Time Series Shapelets. arXiv:1503.03238[cs.LG], March 2015.
  5. Mitzev, I., Younan, N. (2016). Concatenate Decision Paths Classification for Time Series Shapelets. In: CMCA'16, January 2-3, Zurich, Switzerland, 2016.
  6. Mitzev, I., Younan, N. (2015). Time Series Shapelets: Training Time Improvement Based on Particle Swarm Optimization. IJMLC, vol. 5, August 2015
  7. Chen, Y., Keogh, E., Bing, H., Begum, N., Bagnall, A., Mueen, A., Batista, G. (2015). The UCR Time Series Classification Archive. [online]. Available at: http://www.cs.ucr.edu/eamonn/time_series_data/.
  8. Grabocka, J., Wistuba, M., Schmidt-Thieme, L. (2015). Scalable Discovery of Time-Series Shapelets. [online]. Available at: www.dropbox.com/sh/btiee2pyn6a989q/ AACDfzkkpdYPmgw7pgTgUoeYa
  9. Keogh, E., Rakthanmanon, T. (2013). Fast Shapelets: A scalable Algorithm for Discovering Time Series Shapelets. [online]. Available at: http://alumni.cs.ucr.edu/rakthant/FastShapelet
Download


Paper Citation


in Harvard Style

Mitzev I. and H. Younan N. (2017). Concatenated Decision Paths Classification for Datasets with Small Number of Class Labels . In Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-222-6, pages 410-417. DOI: 10.5220/0006190004100417


in Bibtex Style

@conference{icpram17,
author={Ivan Mitzev and Nicolas H. Younan},
title={Concatenated Decision Paths Classification for Datasets with Small Number of Class Labels},
booktitle={Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2017},
pages={410-417},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006190004100417},
isbn={978-989-758-222-6},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 6th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - Concatenated Decision Paths Classification for Datasets with Small Number of Class Labels
SN - 978-989-758-222-6
AU - Mitzev I.
AU - H. Younan N.
PY - 2017
SP - 410
EP - 417
DO - 10.5220/0006190004100417