Comparison of Dimension Reduction Methods for Multivariate Time Series Pattern Recognition

Patrick Petersen, Hanno Stage, Philipp Reis, Jonas Rauch, Eric Sax

2024

Abstract

Large volumes of time series data are frequently analyzed using unsupervised algorithms to identify patterns. Multivariate time series’s time and space complexity poses challenges in this context. Dimensionality reduction, a common technique in data science, provides a viable solution to improve time and space complexity. Nevertheless, a crucial question arises concerning how the time advantage compares to the information loss. This paper compares dimension reduction methods within unsupervised time series pattern recognition, including rule-based, spectral, probabilistic, and unsupervised learning-based approaches. The comparison involves both synthetic and real-world datasets for a comprehensive evaluation. The findings reveal the potential to accelerate pattern recognition algorithms by 90 %, with only 18 % information loss in the sense of the F1 score.

Paper Citation

in Harvard Style

Petersen P., Stage H., Reis P., Rauch J. and Sax E. (2024). Comparison of Dimension Reduction Methods for Multivariate Time Series Pattern Recognition. In Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM; ISBN 978-989-758-684-2, SciTePress, pages 809-816. DOI: 10.5220/0012428900003654

in Bibtex Style

@conference{icpram24,
author={Patrick Petersen and Hanno Stage and Philipp Reis and Jonas Rauch and Eric Sax},
title={Comparison of Dimension Reduction Methods for Multivariate Time Series Pattern Recognition},
booktitle={Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM},
year={2024},
pages={809-816},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012428900003654},
isbn={978-989-758-684-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM
TI - Comparison of Dimension Reduction Methods for Multivariate Time Series Pattern Recognition
SN - 978-989-758-684-2
AU - Petersen P.
AU - Stage H.
AU - Reis P.
AU - Rauch J.
AU - Sax E.
PY - 2024
SP - 809
EP - 816
DO - 10.5220/0012428900003654
PB - SciTePress