ALGORITHM OF HYBRID GMDH-NETWORK CONSTRUCTION FOR TIME SERIES FORECAST

Authors

  • O. I. Chumachenko National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”
  • A. T. Kot National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”
  • А. E. Mandrenko National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”

DOI:

https://doi.org/10.18372/1990-5548.64.14852

Keywords:

Hybrid neural network, structural-parametric synthesis, forecast of time series

Abstract

It is considered the problem of structural-parametric synthesis of a hybrid neural networks based on the use of Group Method of Data Handling neural network. Hybridization is achieved through the use of various neurons: classical, nonlinearAdaline, R-neuron, W-neuron, Wavelet-neuron. The problem of structural-parametric synthesis of hybrid neural network consists in the optimal choice of the number of layers, the number of neurons in the layers, the order of alternation of layers with different neurons. As an example it is considered the forecast problem solution with help of hybrid neural networks based on the data of the COVID-19 pandemic, collected by Johns Hopkins University. A MAPE criterion was used for quality assessment.

Author Biographies

O. I. Chumachenko, National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”

Technical Cybernetic Department

Doctor of Engineering Science. Associate Professor

orcid.org/0000-0003-3006-7460

A. T. Kot, National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”

Technical Cybernetic Department

Post-graduate student

А. E. Mandrenko, National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”

Technical Cybernetic Department

Bachelor

References

V. M. Sineglazov, E. I. Chumachenko, and V. S. Gorbatyuk, Intelligent methods of forecast. Kyiv: Osvita Ukraine, 2013, 236 p.

Ye. Bodyanskiy and O. Vynokurova, “Compartmental adaptive wavelon and its learning algorithm,” Control systems and computers 1, pp. 47–53, 2009.

O. I. Chumachenko, S. T. Dychko and A. R. Rizhiy, “Hybrid Neuron Networks Based on Q-, W- and Classical Neurons,” Electronics and Control Systems, pp. 58–62, 2019. https://doi.org/10.18372/1990-5548.62.14385

Ye. Bodyanskiy, I. Pliss, O. Vynokurova, “Hybrid GMDH-neural network of computational intelligence,” Proc. 3rd International Workshop on Inductive Modelling, Poland, Krynica. 2009, рр. 100–107.

Ye. Bodyanskiy, O. Vynokurova, and N. Teslenko, “Cascade GMDH-Wavelet-Neuro-Fuzzy Network,” The 4th International Workshop on Inductive Modelling IWIM 2011, рр. 22–30.

Е. А. Vinokurova, “Generalized multidimensional wavelet-neuro-fuzzy system in computational intelligence problems,” Intelligent decision-making systems and problems of computational intelligence: a collection of scientific papers based on the materials of an international scientific conference. Evpatoriay–Cherson, vol. 2, 2010, pp. 329–333.

Downloads

Issue

Vol. 2 No. 64 (2020)

Section

COMPUTER SCIENCES AND INFORMATION TECHNOLOGIES

License

Authors who publish with this journal agree to the following terms:

Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.

Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.

Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).