ARTIFICIAL NEURAL NETWORK TRAINING BASED ON PERFORMANCE AND RISKS ASSESSMENT DATA OF THE INVESTMENT IN DIGITAL ASSETS
DOI:
https://doi.org/10.28925/2663-4023.2023.19.135145Keywords:
information Technologies, information systems, data analysis, game theory, fuzzy logic, artificial neural networks, digital assets, cryptocurrencyAbstract
The problem of analyzing the results of training artificial neural networks based on data about the efficiency and risks of investing in digital assets, particularly in the context of managing the buying and selling process of cryptocurrencies, has been investigated. The approach for solving this problem is based on the application of game theory as the main principle for forming the architecture of the artificial neural network. Combining two fundamental theories - game theory and neural networks - allows the creation of intuitively understandable and effective intelligent information systems for decision support in various application areas, such as finance, economics, and resource management. Special attention is paid to considering fuzzy parameters and uncertainties in market conditions, reflecting the real circumstances of investing in cryptocurrencies and other digital assets. The article proposes a series of methods for training and adapting the artificial neural network within the developed approach, as well as recommendations for evaluating its effectiveness and stability. The possible areas of application and prospects for further development of this methodology in the context of the digital asset market have been analyzed. The application of the developed methodology for analyzing the results of artificial neural network training has been illustrated, and its high efficiency in predicting investment performance and risks in digital assets has been confirmed. The issues and limitations that may arise during the use of this methodology were highlighted, and possible ways to overcome and improve them have been proposed..
Downloads
References
Lakhno, V., Akhmetov, B., Malyukov, V., Kartbaev, T. (2018). Modeling of the decision-making procedure for financing of cyber security means of cloud services by the medium of a bilinear multistep quality game with several terminal surfaces. International Journal of Electronics and Telecommunications, 64(4), 467-472.
Z. Liu et al. (2019). A Survey on Blockchain: A Game Theoretical Perspective. IEEE Access, 7, 47615-47643. https://doi.org/10.1109/ACCESS.2019.2909924.
Liu, Z., Luong, N. C., Wang, W., Niyato, D., Wang, P., Liang, Y. C., Kim, D. I. (2019). A survey on applications of game theory in blockchain. arXiv preprint arXiv, 1902,10865.
Bebeshko, B., Malyukov, V., Lakhno, M., Skladannyi, P., Sokolov, V., Shevchenko, S., Zhumadilova, M. (2022) Application of game theory, fuzzy logic and neural networks for assessing risks and forecasting rates of digital currency Journal of Theoretical and Applied Information Technology, 100(24). http://www.jatit.org/volumes/Vol100No24/15Vol100No24.pdf
Trimborn, S., Li, M., Härdle, W. K. (2019). Investing with Cryptocurrencies—a Liquidity Constrained Investment Approach*. Journal of Financial Econometrics, 18(2), 280–306. https://doi.org/10.1093/jjfinec/nbz016
Angerer, M., et al. (2021). Objective and subjective risks of investing into cryptocurrencies. Finance Research Letters , 40, 101737. https://doi.org/10.1016/j.frl.2020.101737
Maiti, M., Vukovic, D., Krakovich, V., Panxy, M. K. (2020). How integrated are cryptocurrencies. International Journal of Big Data Management, 1(1), 64-80. https://doi.org/10.1504/IJBDM.2019.10023285
LSTM-Bitcoin-GoogleTrends-Prediction. https://github.com/falaybeg/LSTM-Bitcoin-GoogleTrends-Prediction
Google Finance. https://www.google.com/finance/quote/BTC-UAH
Buriachok, V., Shevchenko, S., ZhdanovаY., Skladannyi, P. (2021). INTERDISCIPLINARY APPROACH TO THE DEVELOPMENT OF IS RISK MANAGEMENT SKILLS ON THE BASIS OF DECISION-MAKING THEORY. Electronic Professional Scientific Edition «Cybersecurity: Education, Science, Technique», 3(11), 155-165. https://doi.org/10.28925/2663-4023.2021.11.155165
