Control Systems and Computers

https://doi.org/10.15407/csc.2023.03.069

Control Systems and Computers, 2023, Issue 3 (303), pp. 69-76.

 UDK 004.4

A.M. Holiachenko, PhD Student at the Department of Computer Systems Software, Faculty of Applied Mathematics, National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, ORCID: https://orcid.org/0009-0008-5715-3805, Prosp. Beresteiskyi, 37, Kyiv 03056, Ukraine,
anastasiia.holiachenko@gmail.com 

L.A. Lіushenko, PhD in Technical Sciences, associate professor of Computer Systems Software Department  National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, ORCID: https://orcid.org/0000-0003-4319-5955, Scopus Author ID: 57208345053, Prosp. Beresteiskyi, 37, Kyiv 03056, Ukraine, lyushenkol@gmail.com

SOFTWARE EXPERT SYSTEM FOR CHOOSING CRYPTOCURRENCY FORECASTING ALGORITHMS IN REAL-TIME

Introduction. Today there is a large number of cryptocurrencies in the world with their own unique characteristics. Traders and investors working in the cryptocurrency market use various software to analyze and forecast the exchange rate of cryptocurrencies. The ability to correctly and quickly make decisions using the results of cryptocurrency rate forecasting is currently defined as a key goal for traders, investors and analysts of this market.

Purpose. In this article, the existing methods of analysis and forecasting of the exchange rate on cryptocurrency exchanges were considered, namely – fundamental and technical analysis, machine learning, news flow analysis, and a hybrid approach.

During the study, it was demonstrated that different algorithms can have a high level of probability of predictions in a specific situation and with optimized parameters, but at the same time show much lower probability indicators with the slightest change in parameters or dynamics of the cryptocurrency exchange rate. Accordingly, a perfect method with universal parameters that will always consistently show high results in the probability of predictions does not exist.

The implementation of a software automatic expert system for choosing algorithms in real-time will allow automatically choose the best algorithm for cryptocurrency forecasting based on the analysis of the effectiveness of algorithms over the last N iterations in the past.

Methods. By the development of automatic expert system for choosing cryptocurrency forecasting algorithms in real-time.

Results. The process of analyzing and forecasting the cryptocurrency exchange rates with an existing software expert system for choosing algorithms for cryptocurrency forecasting is considered, and the logic of the operation of such an expert system is presented. Also, ways to solve problems that may arise during the application of this system were identified and substantiated.

Conclusion. The alternative approach to the analysis and forecasting of cryptocurrencies in real-time was implemented in the software expert system for choosing alghorithms, which allows stabilizing the high probability of the exchange rate forecasts on the cryptocurrency market in comparison with the use of only one method or the hybridization of several methods.

Download full text! (On English)

Keywords: cryptocurrency exchange rate forecasting, trading, forecasting algorithms, time series analysis, expert system, technical analysis, cryptocurrency market.

1. Achelis, S.B. (2000). Technical Analysis from A to Z, 2nd Edition. McGraw-Hill.
2. Holiachenko, A., Lyushenko, L., Strutsynsky, O. (2022). “Modified Method of Cryptocurrency Exchange Rate Forecasting Based on ARIMA Class Models with Data Verification”. Advances in Artificial Systems for Logistics Engineering. Lecture Notes on Data Engineering and Communications Technologies, vol 135, pp 123 -136.
   https://doi.org/10.1007/978-3-031-04809-8_11
3. Lyushenko, L., Holiachenko, A. (2020). “Optimization of the Method of Technical Analysis of Cryptocurrency Price Differences Movements”. Advances in Computer Science for Engineering and Education II. ICCSEEA 2019. Advances in Intelligent Systems and Computing, vol 938, pp 388-397.
   https://doi.org/10.1007/978-3-030-16621-2_36
4. Breiman, L. (2001). “Random Forests”. Machine Learning, 45, pp 5 -32.
   https://doi.org/10.1023/A:1010933404324
5. Burkov, A. (2019). The Hundred-Page Machine Learning Book. Andriy Burkov.
6. Palepu, K.G., Healy, P.M. (2008).Business Analysis & Valuation: Using Financial Statements, 4th ed. Mason OH: Thomson South-Western.
7. McKinsey and Company, Koller, T., Goedhart, M., Wessels, D. (2020). Valuation: Measuring and Managing the Value of Companies. Seventh ed. Hoboken New Jersey: John Wiley & Sons.
8. Coulter, K.A. (2022). “The impact of news media on Bitcoin prices: modelling data driven discourses in the crypto-economy with natural language processing”. R Soc Open Sci, 9(4).
   https://doi.org/10.1098/rsos.220276
9. Tran, T. (2022). Predicting Digital Asset Prices using Natural Language Processing: a survey.
10. Bishop, C.M. (2006). “Pattern Recognition and Machine Learning”. Information Science and Statistics. New York: Springer-Verlag, pp. 653 -673.
11. Valipour, M. (2015). “Long-term runoff study using SARIMA and ARIMA models in the United States”. Meteorological Applications. https://doi.org/10.1002/met.1491
12. Akyildirim, E., Göncü, A., Sensoy, A. (2018). Prediction of Cryptocurrency Returns using Machine Learning.
13. Kim, K.-J. (2003). “Financial time series forecasting using support vector machines”, Neurocomputing, vol 55, pp. 307-319.
    https://doi.org/10.1016/S0925-2312(03)00372-2
14. Lee, M.-C. (2009). “Using support vector machine with a hybrid feature selection method to the stock trend prediction”. Expert Systems with Applications, 36(8), pp. 10896-10904.
    https://doi.org/10.1016/j.eswa.2009.02.038
15. Kumar, M., Thenmozhi, M. (2006). “Forecasting Stock Index Movement: A Comparison of Support Vector Machines and Random Forest”. Indian Institute of Capital Markets 9th Capital Markets Conference Paper. .
    https://doi.org/10.2139/ssrn.876544
16. Giarratano J., Riley G. (1998). Expert Systems: Principles and Programming. 3rd ed. Boston: PWS Pub.
17. Harmon, P., King, D.R. (1985). Expert systems: artificial intelligence in business. Technology and Culture, 28, 894 p. https://doi.org/10.2307/3105218
18. Guresen, E., Kayakutlu, G., Daim, T. (2011). “Using artificial neural network models in stock market index prediction”. Expert Systems with Applications, 38, pp. 10389 -10397.
    https://doi.org/10.1016/j.eswa.2011.02.068

Received 18.09.2023