Prediction of complex processes by neural networks

  • Alfira M. Kumratova Kuban State Agrarian University
  • Kirill E. Chumachenko Kuban State Agrarian University
Keywords: artificial neural networks, forecast, time series, data analysis, big data, predictive modeling, CNN, lstm, forecasting, complex process

Abstract

Importance: forecasting of complex processes is necessary in various industries to optimize their operations, reduce costs and increase efficiency. Traditional statistical methods and machine learning algorithms were used for forecasting, but the advent of neural networks significantly increased the accuracy of forecasts. Purpose: the use of neural networks for predicting complex processes is becoming increasingly popular due to their ability to study complex relationships between data, find patterns in them and generalize them, predicting future results. Research design: аssuming that the use of neural networks for time series forecasting is associated with the problems of developing accurate models, the paper shows the potential advantages of using neural networks for predictive modeling of complex processes. Results: the authors present a demonstration of the work of a tool for predicting time series using neural networks.

Downloads

Download data is not yet available.

Author Biographies

Alfira M. Kumratova , Kuban State Agrarian University

Cand. Sci. (Econ.)

Kirill E. Chumachenko , Kuban State Agrarian University

student

References

Егоров И.А. Методы анализа Big Data / И.А. Егоров, Г.С. Аматунянц, А.М. Кумратова // Цифровизация экономики: направления, методы, инструменты: Сборник материалов v Всероссийской научно-практической конференции, Краснодар, 16–21 января 2023 года. Краснодар, Кубанский государственный аграрный университет имени И.Т. Трубилина, 2023, с. 192-196.

Манжула В.Г. Нейронные сети Кохонена и нечеткие нейронные сети в интеллектуальном анализе данных / В.Г. Манжула, Д.С. Федяшов // Фундаментальные исследования, 2011, no. 4, с. 108-114.

Обзор современных моделей и методов анализа временных рядов динамики процессов в социальных, экономических и социотехнических системах / Е.Г. Андрианова, С.А. Головин, С.В. Зыков [и др.] // Российский технологический журнал, 2020, т. 8, no. 4(36), с. 7-45.

Шугай Ю.С. Нейросетевые алгоритмы прогнозирования событий и поиска предвестников в многомерных временных рядах // Искусственный интеллект, 2004, no. 2, с. 211-215.

Якимова В.А. Возможности и перспективы использования цифровых технологий в аудиторской деятельности // Вестник Санкт-Петербургского университета. Экономика, 2020, т. 36, no. 2, с. 287-318.

Baldwin A. Opportunities for artificial intelligence development in the accounting domain: the case for auditing / A. Baldwin, C. Brown, B. Trinkle // Finance and Management, 2006, no. 14, рр. 77-86.

Calderon T.G. A roadmap for future neural research in auditing and risk assessment / T.G. Calderon, J.J. Cheh // International Journal of Accounting Information Systems, 2002, vol. 3-4, pp. 203-236.

Chiu C.T. An intelligent forecasting support system in auditing: expert system and neural network approach / C.T. Chiu, R. Scott // System Sciences, 1994, vol. 3, pp. 272-280.

Preliminary rocessing of data for the problem of short-term forecasting of electric energy cost / A.S. Polyakova, U.N. Kruglova, S.S. Bezhitskiy, S.V. Shelikhova // Молодежь. Общество. Современная наука, техника и инновации, 2015, no. 14, рр. 247-249.

Sermpinis G. Forecasting and trading the EUR/USD exchange rate with gene expression and psi sigma neural networks / G. Sermpinis, J. Laws, A. Karathanasopoulos, C.L. Dunis // Expert Sys. Appl., 2012, no. 39(10), рр. 8865-8877.

Vanstone B. An empirical methodology for developing stockmarket trading systems using artificial neural networks / B. Vanstone, G. Finnie // Expert Sys. Appl., 2009, no. 36(3), рр. 6668-6680.

Xianjun Ni. Research of Data Mining Based on Neural Networks / Ni Xianjun // World Academy of Science, Engineering and Technology, 2008, no. 39, рр. 381384.

Published
2023-05-15
How to Cite
Kumratova , A. M., & Chumachenko , K. E. (2023). Prediction of complex processes by neural networks. Modern Economics: Problems and Solutions, 3, 27-36. Retrieved from https://journals.vsu.ru/meps/article/view/11235
Section
Mathematical and Instrumental Methods in Economics