Two approaches to machine learning classification of time series based on recurrence plots

Lyudmyla Kirichenko; Tamara Radivilova; Vitalii Bulakh; Petro Zinchenko; Abed Saif Alghawli

doi:10.1109/DSMP47368.2020.9204021

Two approaches to machine learning classification of time series based on recurrence plots

Lyudmyla Kirichenko, Tamara Radivilova, Vitalii Bulakh, Petro Zinchenko, Abed Saif Alghawli

Computer Sciences

Kharkiv National University of Radio Electronics

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

13 Scopus citations

Abstract

The article considers the task of classifying fractal time series based on the construction of their recurrence plots. Short realizations of EEG signals were used as input data. Two classification machine learning methods were considered: in the first case, quantitative fractal and recurrent characteristics of the time series were classification features, in the second case, image recognition of recurrence plots was carried out. The results showed fairly high classification quality for both methods, and relative advantage of the image classification method.

Original language	English
Title of host publication	Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	84-89
Number of pages	6
ISBN (Electronic)	9781728132143
DOIs	https://doi.org/10.1109/DSMP47368.2020.9204021
State	Published - Aug 2020
Event	3rd IEEE International Conference on Data Stream Mining and Processing, DSMP 2020 - Lviv, Ukraine Duration: 21 Aug 2020 → 25 Aug 2020

Publication series

Name	Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020

Conference

Conference	3rd IEEE International Conference on Data Stream Mining and Processing, DSMP 2020
Country/Territory	Ukraine
City	Lviv
Period	21/08/20 → 25/08/20

Keywords

Classification of time series
Fractal time series
Machine learning classification
Recurrence plot

Access to Document

10.1109/DSMP47368.2020.9204021

Cite this

Kirichenko, L., Radivilova, T., Bulakh, V., Zinchenko, P., & Saif Alghawli, A. (2020). Two approaches to machine learning classification of time series based on recurrence plots. In Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020 (pp. 84-89). Article 9204021 (Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/DSMP47368.2020.9204021

Kirichenko, Lyudmyla ; Radivilova, Tamara ; Bulakh, Vitalii et al. / Two approaches to machine learning classification of time series based on recurrence plots. Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020. Institute of Electrical and Electronics Engineers Inc., 2020. pp. 84-89 (Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020).

@inproceedings{83b9c50311554a3ebd4cd437dabcefdd,

title = "Two approaches to machine learning classification of time series based on recurrence plots",

abstract = "The article considers the task of classifying fractal time series based on the construction of their recurrence plots. Short realizations of EEG signals were used as input data. Two classification machine learning methods were considered: in the first case, quantitative fractal and recurrent characteristics of the time series were classification features, in the second case, image recognition of recurrence plots was carried out. The results showed fairly high classification quality for both methods, and relative advantage of the image classification method.",

keywords = "Classification of time series, Fractal time series, Machine learning classification, Recurrence plot",

author = "Lyudmyla Kirichenko and Tamara Radivilova and Vitalii Bulakh and Petro Zinchenko and \{Saif Alghawli\}, Abed",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 3rd IEEE International Conference on Data Stream Mining and Processing, DSMP 2020 ; Conference date: 21-08-2020 Through 25-08-2020",

year = "2020",

month = aug,

doi = "10.1109/DSMP47368.2020.9204021",

language = "English",

series = "Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "84--89",

booktitle = "Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020",

address = "United States",

}

Kirichenko, L, Radivilova, T, Bulakh, V, Zinchenko, P & Saif Alghawli, A 2020, Two approaches to machine learning classification of time series based on recurrence plots. in Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020., 9204021, Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020, Institute of Electrical and Electronics Engineers Inc., pp. 84-89, 3rd IEEE International Conference on Data Stream Mining and Processing, DSMP 2020, Lviv, Ukraine, 21/08/20. https://doi.org/10.1109/DSMP47368.2020.9204021

Two approaches to machine learning classification of time series based on recurrence plots. / Kirichenko, Lyudmyla; Radivilova, Tamara; Bulakh, Vitalii et al.
Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020. Institute of Electrical and Electronics Engineers Inc., 2020. p. 84-89 9204021 (Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Two approaches to machine learning classification of time series based on recurrence plots

AU - Kirichenko, Lyudmyla

AU - Radivilova, Tamara

AU - Bulakh, Vitalii

AU - Zinchenko, Petro

AU - Saif Alghawli, Abed

PY - 2020/8

Y1 - 2020/8

N2 - The article considers the task of classifying fractal time series based on the construction of their recurrence plots. Short realizations of EEG signals were used as input data. Two classification machine learning methods were considered: in the first case, quantitative fractal and recurrent characteristics of the time series were classification features, in the second case, image recognition of recurrence plots was carried out. The results showed fairly high classification quality for both methods, and relative advantage of the image classification method.

AB - The article considers the task of classifying fractal time series based on the construction of their recurrence plots. Short realizations of EEG signals were used as input data. Two classification machine learning methods were considered: in the first case, quantitative fractal and recurrent characteristics of the time series were classification features, in the second case, image recognition of recurrence plots was carried out. The results showed fairly high classification quality for both methods, and relative advantage of the image classification method.

KW - Classification of time series

KW - Fractal time series

KW - Machine learning classification

KW - Recurrence plot

UR - http://www.scopus.com/inward/record.url?scp=85093649952&partnerID=8YFLogxK

U2 - 10.1109/DSMP47368.2020.9204021

DO - 10.1109/DSMP47368.2020.9204021

M3 - Conference contribution

AN - SCOPUS:85093649952

T3 - Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020

SP - 84

EP - 89

BT - Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 3rd IEEE International Conference on Data Stream Mining and Processing, DSMP 2020

Y2 - 21 August 2020 through 25 August 2020

ER -

Kirichenko L, Radivilova T, Bulakh V, Zinchenko P, Saif Alghawli A. Two approaches to machine learning classification of time series based on recurrence plots. In Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020. Institute of Electrical and Electronics Engineers Inc. 2020. p. 84-89. 9204021. (Proceedings of the 2020 IEEE 3rd International Conference on Data Stream Mining and Processing, DSMP 2020). doi: 10.1109/DSMP47368.2020.9204021

Two approaches to machine learning classification of time series based on recurrence plots

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this