vishalbelsare / dba

DBA: Averaging for Dynamic Time Warping

DBA

DBA stands for Dynamic Time Warping Barycenter Averaging. DBA is an averaging method that is consistent with Dynamic Time Warping. I give below an example of the difference between the traditional arithmetic mean of the set of time series and DBA.

Underlying research and scientific papers

This code is supporting 3 research papers:

• Pattern Recognition 2011: A global averaging method for Dynamic Time Warping
• ICDM 2014: Dynamic Time Warping Averaging of Time Series allows Faster and more Accurate Classification
• ICDM 2017: Generating synthetic time series to augment sparse datasets

When using this repository, please cite:

@ARTICLE{Petitjean2011-DBA,
  title={A global averaging method for dynamic time warping, with applications to clustering},
  author={Petitjean, Fran{\c{c}}ois and Ketterlin, Alain and Gan{\c{c}}arski, Pierre},
  journal={Pattern Recognition},
  volume={44},
  number={3},
  pages={678--693},
  year={2011},
  publisher={Elsevier}
}

@INPROCEEDINGS{Petitjean2014-ICDM-2,
  title={Dynamic time warping averaging of time series allows faster and more accurate classification},
  author={Petitjean, Fran{\c{c}}ois and Forestier, Germain and Webb, Geoffrey I and Nicholson, Ann E and Chen, Yanping and Keogh, Eamonn},
  booktitle={Data Mining (ICDM), 2014 IEEE International Conference on},
  pages={470--479},
  year={2014},
  organization={IEEE}
}

@INPROCEEDINGS{Forestier2017-ICDM,
  title={Generating synthetic time series to augment sparse datasets},
  author={Forestier, Germain and Petitjean, Fran{\c{c}}ois and Dau, Hoang Anh and Webb, Geoffrey I and Keogh, Eamonn},
  booktitle={Data Mining (ICDM), 2017 IEEE International Conference on},
  pages={865--870},
  year={2017},
  organization={IEEE}
}

Organisation of the repository

This repository gives you different versions of DBA for different programming language, whether you want to have a warping window or not, etc. Apologies for the inconsistencies between versions but I've basically created them as the need arose.

Each file corresponds to one of these combinations; if one is missing for your usage, let me know. Currently the length is limited to 1,000 (let me know if you need more).

• DBA.java standard DBA in Java with no warping window and memory allocated statically
• DBAWarpingWindow.java same as DBA.java but with a warping window as a parameter
• DBA.m Matlab implementation of DBA with no windows
• DBA.py Fast Python implementation of DBA with no windows
• DBA_multivariate.py Fast Python implementation of DBA for multi-variate time series with no windows
• cython/* Cython implementation (thus usable in Python) of DBA with warping window (mono-variate)