ARTIFICIAL NEURAL NETWORKS FOR MODELING OF CRISIS MANAGEMENT OF NATIONAL ECONOMY

N.Ya. Savka

Èlektron. model. 2020, 42(2):109-120
https://doi.org/10.15407/emodel.42.02.109

ABSTRACT

The artificial neural networks with radial basis functions has been analyzed as the most effective for modeling processes with deep instability. The algorithm of tuning parameters of artificial radial-type neural networks has been described as well as the priorities of the crisis management of the national economy. The optimal architecture of artificial neural networks has been developed, the basic functions of which are radial for the crisis management of the national economy system modeling. The result of modeling of crisis management indicators, based on the developed architecture of artificial neural network, has been presented. The efficiency of using artificial radial-type neural networks for crisis prevention has been investigated.

KEYWORDS

artificial neural networks, radial basis functions, crisis management, modeling.

REFERENCES

  1. Malyi, I., Radionova, I. and Yemelianenko, L. et al. (2017), Antykryzove upravlinnia natsionalnoiu ekonomikoiu [Crisis management of the national economy], KNEU, Kyiv, Ukraine.
  2. Klebanova, T.S., Dymchenko, O.V., Rudachenko, O.O. and Hvozdytskyi, V.S. (2018), Neiromerezhevi modeli otsinky finansovykh kryz na pidpryiemstvakh korporatyvnoho typu [Neural network models of financial crisis assessment at enterprises of corporate type], KHNUMH im. O.M. Beketova, Kharkiv, Ukraine.
  3. Klebanova, T.S, Hrachev, V.I., Raevneva, E.V., Hurianova, L.S. and Poliakova, O.Ya. (2007), Mehanizmyi i modeli upravleniya krizisnyimi situatsiyami [Mechanisms and models of crisis management], INGEC.
  4. Ivanets, O.B., Bukrieieva, O.V. and Dvornik, M.V. (2011) «Construction of prediction models using  artificial  neural  networks»,  Electronics  and  control  systems,   4(30), рр. 139-142.
    https://doi.org/10.18372/1990-5548.30.922
  5. Bodyanskіy, E.V. and Rudenko, O. H. (2004), Ickucctvennyie neyronnyie ceti: arhitekturyi, obuchenie, primeneniya [Artificial neural networks: architectures, training, applications], TELETECH, Kharkiv, Ukraine.
  6. Kalinina, I.O. (2009), “Investigation of neural network learning algorithms in forecasting tasks”, Scientific papers, Vol. 104, Iss. 117, рр. 160-171.
  7. Nelles, O. (2001), Nonlinear Systems Identification, Springer, Berlin.
    https://doi.org/10.1007/978-3-662-04323-3
  8. Savka, N.Ya., Spilchuk, V.M. and Spivak, I.Ya (2010) “Problems of identification of artificial neural networks with radial basis functions and possible directions of their solution”, Inductive modeling of complex systems, Vol. 2, pp. 181-193.
  9. Shtovba, S.D. (2001) “Introduction to fuzzy set theory and fuzzy logic”, available at: http://matlab.exponenta.ru/fuzzylogic/book1/index.php (accessed February 19, 2020).
  10. Dyvak, М., Maslyiak, Y., Papa, O. and Savka, N. (2017) “Clustering and interval analysis of heterogeneous data sample”, Proceeding of 12th International Conference Computer Sciences and Information Technologies (CSIT), pp. 528-532.
    https://doi.org/10.1109/STC-CSIT.2017.8098843

Full text: PDF