- Physical Review Journals
All JournalsPhysics Magazine

Physical Review Research

Reuse & Permissions

It is not necessary to obtain permission to reuse this article or its components as it is available under the terms of the Creative Commons Attribution 4.0 International license. This license permits unrestricted use, distribution, and reproduction in any medium, provided attribution to the author(s) and the published article's title, journal citation, and DOI are maintained. Please note that some figures may have been included with permission from other third parties. It is your responsibility to obtain the proper permission from the rights holder directly for these figures.

Export citation

Export citation

Choose format for download:

Download Citation
  • Open Access

Comparison of the performance of features based on symbolic dynamics and recurrence microstates for time series classification

H. C. Costa1,2, J. V. M. Silveira1,2, B. R. R. Boaretto3,4, C. Masoller3, S. R. Lopes1,2, and T. L. Prado1,2,*

  • *Contact author: thiago@fisica.ufpr.br
PDF

Phys. Rev. Research 8, 023165 – Published 14 May, 2026

DOI: https://doi.org/10.1103/1m3f-jshf

Abstract

Symbolization is a well-established tool to transform a time series into a sequence of symbols. Here, we compare two symbolization approaches: recurrence microstate (RM), based on recurrence patterns in phase space, and symbolic dynamics (SD), derived from ordinal binary sequences. Specifically, we compare the performance of two sets of features, based on RM and SD, for parameter estimation and classification. For parameter estimation, we generate synthetic data simulated from chaotic dynamical systems with a parameter that varies in a range that is divided into ten equal bins, and for classification, we use freely available electrocardiogram (ECG) recordings from patients with four different cardiac pathologies. We train a Random Forest (RF) classifier to infer to which parameter bin a synthetic time series belongs and to which pathology an ECG recording corresponds. We find that using RM features to train the RF algorithm consistently provides higher accuracy; however, SD features encapsulate complementary information that can improve the performance. Combining the two sets of features generally enhances the performance.

View figure in article

Physics Subject Headings (PhySH)

Article Text

References (42)

  1. E. Ott, Chaos in Dynamical Systems (Cambridge University Press, Cambridge, 2002).
  2. G. Datseris and U. Parlitz, Nonlinear Dynamics: A Concise Introduction Interlaced with Code, Undergraduate Lecture Notes in Physics (Springer International Publishing, Cham, 2022).
  3. H. Kantz and T. Schreiber, Nonlinear Time Series Analysis (Cambridge University Press, Cambridge, 2004).
  4. C. Bandt and B. Pompe, Permutation entropy: A natural complexity measure for time series, Phys. Rev. Lett. 88, 174102 (2002).
  5. N. Marwan, M. C. Romano, M. Thiel, and J. Kurths, Recurrence plots for the analysis of complex systems, Phys. Rep. 438, 237 (2007).
  6. M. Barreiro, A. C. Marti, and C. Masoller, Inferring long memory processes in the climate network via ordinal pattern analysis, Chaos 21, 013101 (2011).
  7. J.-P. Eckmann, S. O. Kamphorst, and D. Ruelle, Recurrence plots of dynamical systems, Europhys. Lett. 4, 973 (1987).
  8. H. Poincaré, Sur le probleme des trois corps et les equations de la dynamique, Acta Mathematica (F. & G. Beijer, Stockholm, 1890).
  9. G. Corso, T. d. L. Prado, G. Z. d. S. Lima, J. Kurths, and S. R. Lopes, Quantifying entropy using recurrence matrix microstates, Chaos 28, 083108 (2018).
  10. S. L. Brunton and J. N. Kutz, Data-Driven Science and Engineering: Machine Learning, Dynamical Systems, and Control (Cambridge University Press, Cambridge, 2019).
  11. T. L. Prado, B. R. R. Boaretto, G. Corso, G. Z. dos Santos Lima, J. Kurths, and S. R. Lopes, A direct method to detect deterministic and stochastic properties of data, New J. Phys. 24, 033027 (2022).
  12. G. S. Spezzatto, J. V. V. Flauzino, G. Corso, B. R. R. Boaretto, E. E. N. Macau, T. L. Prado, and S. R. Lopes, Recurrence microstates for machine learning classification, Chaos 34, 073140 (2024).
  13. P. M. Addo, M. Billio, and D. Guegan, Nonlinear dynamics and recurrence plots for detecting financial crisis, North Am. J. Econ. Finance 26, 416 (2013).
  14. O. A. Rosso, H. A. Larrondo, M. T. Martin, A. Plastino, and M. A. Fuentes, Distinguishing noise from chaos, Phys. Rev. Lett. 99, 154102 (2007).
  15. J. M. Amigó, S. Zambrano, and M. A. F. Sanjuán, Permutation complexity of spatiotemporal dynamics, Europhys. Lett. 90, 10007 (2010).
  16. B. R. R. Boaretto, R. C. Budzinski, K. L. Rossi, T. L. Prado, S. R. Lopes, and C. Masoller, Discriminating chaotic and stochastic time series using permutation entropy and artificial neural networks, Sci. Rep. 11, 15789 (2021).
  17. T. L. Prado, G. Corso, G. Z. dos Santos Lima, R. C. Budzinski, B. R. R. Boaretto, F. A. S. Ferrari, E. E. N. Macau, and S. R. Lopes, Maximum entropy principle in recurrence plot analysis on stochastic and chaotic systems, Chaos 30, 043123 (2020).
  18. J. F. Donges, Y. Zou, N. Marwan, and J. Kurths, The backbone of the climate network, Europhys. Lett. 87, 48007 (2009).
  19. N. Marwan, J. F. Donges, Y. Zou, R. V. Donner, and J. Kurths, Complex network approach for recurrence analysis of time series, Phys. Lett. A 373, 4246 (2009).
  20. J. P. Zbilut, N. Thomasson, and C. L. Webber, Recurrence quantification analysis as a tool for nonlinear exploration of nonstationary cardiac signals, Med. Eng. Phys. 24, 53 (2002).
  21. C. Letellier, H. Rabarimanantsoa, L. Achour, A. Cuvelier, and J. F. Muir, Recurrence plots for dynamical analysis of non-invasive mechanical ventilation, Philos. Trans. R. Soc. A 366, 621 (2008).
  22. A. Fabretti and M. Ausloos, Recurrence plot and recurrence quantification analysis techniques for detecting a critical regime, Int. J. Mod. Phys. C 16, 671 (2005).
  23. L. L. Trulla, A. Giuliani, J. P. Zbilut, and C. L. Webber, Recurrence quantification analysis of the logistic equation with transients, Phys. Lett. A 223, 255 (1996).
  24. N. Marwan, N. Wessel, U. Meyerfeldt, A. Schirdewan, and J. Kurths, Recurrence-plot-based measures of complexity and their application to heart-rate-variability data, Phys. Rev. E 66, 026702 (2002).
  25. N. Marwan and K. H. Kraemer, Trends in recurrence analysis of dynamical systems, Eur. Phys. J.: Spec. Top. 232, 5 (2023).
  26. T. L. Prado, V. S. Machado, G. Corso, G. Z. Dos Santos Lima, and S. R. Lopes, How to compute suitable vicinity parameter and sampling time of recurrence analysis, Nonlinear Dyn. 112, 1141 (2024).
  27. A. L. Goldberger, L. A. N. Amaral, L. Glass, J. M. Hausdorff, P. C. Ivanov, R. G. Mark, J. E. Mietus, G. B. Moody, C.-K. Peng and H. E. Stanley, Physiobank, physiotoolkit, and physionet: Components of a new research resource for complex physiologic signals, Circulation 101, e215 (2000).
  28. G. Datseris, Dynamicalsystems.jl: A julia software library for chaos and nonlinear dynamics, J. Open Source Software 3, 598 (2018).
  29. F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay, Scikit-learn: Machine learning in Python, J. Mach. Learn. Res. 12, 2825 (2011).
  30. M. Hénon, A two-dimensional mapping with a strange attractor, Commun. Math. Phys. 50, 69 (1976).
  31. K. Ikeda, Multiple-valued stationary state and its instability of the transmitted light by a ring cavity system, Opt. Commun. 30, 257 (1979).
  32. E. N. Lorenz, Deterministic nonperiodic flow, J. Atmos. Sci. 20, 130 (1963).
  33. O. E. Rössler, An equation for continuous chaos, Phys. Lett. A 57, 397 (1976).
  34. A. M. Fraser and H. L. Swinney, Independent coordinates for strange attractors from mutual information, Phys. Rev. A 33, 1134 (1986).
  35. J. Zheng, H. Guo, and H. Chu, A large scale 12-lead electrocardiogram database for arrhythmia study, PhysioNet (2022), doi:10.13026/wgex-er52.
  36. J. Zheng, H. Chu, D. Struppa, J. Zhang, S. M. Yacoub, H. El-Askary, A. Chang, L. Ehwerhemuepha, I. Abudayyeh, A. S. Barrett, G. Fu, H. Yao, D. Li, H. Guo, and C. Rakovski, Optimal multi-stage arrhythmia classification approach, Sci. Rep. 10, 2898 (2020).
  37. B. R. R. Boaretto, A. C. Andreani, S. R. Lopes, T. L. Prado, and E. E. N. Macau, The use of entropy of recurrence microstates and artificial intelligence to detect cardiac arrhythmia in ECG records, Appl. Math. Comput. 475, 128738 (2024).
  38. O. Yildirim, M. Talo, E. J. Ciaccio, R. S. San Tan, and U. R. Acharya, Accurate deep neural network model to detect cardiac arrhythmia on more than 10,000 individual subject ECG records, Comput. Methods Programs Biomed. 197, 105740 (2020).
  39. Z. Li, D. Zhou, L. Wan, J. Li, and W. Mou, Heartbeat classification using deep residual convolutional neural network from 2-lead electrocardiogram, J. Electrocardiol. 58, 105 (2020).
  40. U. R. Acharya, O. Faust, V. Sree, G. Swapna, R. J. Martis, N. A. Kadri, and J. S. Suri, Linear and nonlinear analysis of normal and cad-affected heart rate signals, Comput. Methods Programs Biomed. 113, 55 (2014).
  41. R. J. Martis, U. R. Acharya, H. Prasad, C. K. Chua, C. M. Lim, and J. S. Suri, Application of higher order statistics for atrial arrhythmia classification, Biomed. Signal Process. Control 8, 888 (2013).
  42. https://github.com/JorgeMalosti/RM-SD-Data.

Outline

Information

Phys. Rev. Research 8, 023165– Published 14 May, 2026
  • Received 2 September 2025
  • Accepted 17 March 2026
Crossmark - Check for updates button
Creative Commons Logo

Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article's title, journal citation, and DOI.

Published by the American Physical Society

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation