Gramian Angular and Markov Transition Fields applied to Time Series Ordinal Classification

Authors

Víctor Manuel Vargas-Yun

Rafael Ayllón-Gavilán

Antonio Manuel Durán-Rosal

Pedro Antonio Gutiérrez

César Hervás-Martínez

David Guijo-Rubio

Conference Proceedings

IWANN 2023: Advances in computational intelligence

Vol. 14135

Pages 505–516

Abstract

This work presents a novel ordinal Deep Learning (DL) approach to Time Series Ordinal Classification (TSOC) field. TSOC consists in classifying time series with labels showing a natural order between them. This particular property of the output variable should be exploited to boost the performance for a given problem. This paper presents a novel DL approach in which time series are encoded as 3-channels images using Gramian Angular Field and Markov Transition Field. A soft labelling approach, which considers the probabilities generated by a unimodal distribution for obtaining soft labels that replace crisp labels in the loss function, is applied to a ResNet18 model. Specifically, beta and triangular distributions have been applied. They have been compared against three state-of-the-art deep learners in the Time Series Classification (TSC) field using 13 univariate and multivariate time series datasets. The approach considering the triangular distribution (O-GAMTFT) outperforms all the techniques benchmarked.

Keywords

gramarian angular fields markov transition fields soft labelling time series ordinal classification

BibTex Citation

@inproceedings{Vargas2023Gramian,
	author = {Vargas-Yun, V{\' i}ctor Manuel and Ayll{\' o}n-Gavil{\' a}n, Rafael and Dur{\' a}n-Rosal, Antonio Manuel and Guti{\' e}rrez, Pedro Antonio and Herv{\' a}s-Mart{\' i}nez, C{\' e}sar and Guijo-Rubio, David},
	booktitle = {IWANN 2023: Advances in computational intelligence},
	doi = {10.1007/978-3-031-43078-7_41},
	year = {2023},
	pages = {505--516},
	organization = {Springer},
	title = {Gramian {Angular} and {Markov} {Transition} {Fields} applied to {Time} {Series} {Ordinal} {Classification}},
	url = {https://link.springer.com/chapter/10.1007/978-3-031-43078-7_41},
	volume = {14135},
}

BibTex Unicode Citation

@inproceedings{Vargas2023Gramian,
	author = {Vargas-Yun, Víctor Manuel and Ayllón-Gavilán, Rafael and Durán-Rosal, Antonio Manuel and Gutiérrez, Pedro Antonio and Hervás-Martínez, César and Guijo-Rubio, David},
	booktitle = {IWANN 2023: Advances in computational intelligence},
	doi = {10.1007/978-3-031-43078-7_41},
	year = {2023},
	pages = {505--516},
	organization = {Springer},
	title = {Gramian {Angular} and {Markov} {Transition} {Fields} applied to {Time} {Series} {Ordinal} {Classification}},
	url = {https://link.springer.com/chapter/10.1007/978-3-031-43078-7_41},
	volume = {14135},
}

APA Citation

Vargas-Yun, V. M., Ayllón-Gavilán, R., Durán-Rosal, A. M., Gutiérrez, P. A., Hervás-Martínez, C., & Guijo-Rubio, D. (2023). Gramian Angular and Markov Transition Fields applied to Time Series Ordinal Classification. IWANN 2023: Advances in Computational Intelligence, 14135, 505–516. https://doi.org/10.1007/978-3-031-43078-7_41

RIS Citation

TY  - CONF
AU  - Vargas-Yun, Víctor Manuel
AU  - Ayllón-Gavilán, Rafael
AU  - Durán-Rosal, Antonio Manuel
AU  - Gutiérrez, Pedro Antonio
AU  - Hervás-Martínez, César
AU  - Guijo-Rubio, David
C3  - IWANN 2023: Advances in computational intelligence
DA  - 2023///
C2  - 2023
DO  - 10.1007/978-3-031-43078-7_41
ID  - temp_id_987809929667
PB  - Springer
SP  - 505-516
TI  - Gramian Angular and Markov Transition Fields applied to Time Series Or
dinal Classification
UR  - https://link.springer.com/chapter/10.1007/978-3-031-43078-7_41
VL  - 14135
ER  -

CV Citation

V.M. Vargas-Yun, R. Ayllón-Gavilán (CA), A.M. Durán-Rosal, P.A. Gutiérrez, C. Hervás-Martínez, D. Guijo-Rubio (1/6). "Gramian Angular and Markov Transition Fields applied to Time Series Ordinal Classification". IWANN 2023: Advances in computational intelligence, pp. 505–516, 2023.