BUSINESS INFORM JOURNAL.

УКР

ENG

Search:

REGISTRATION CERTIFICATE

KV #19905-9705 PR dated 02.04.2013.

FOUNDERS

RESEARCH CENTRE FOR INDUSTRIAL DEVELOPMENT PROBLEMS of NAS (KHARKIV, UKRAINE)

According to the decision No. 802 of the National Council of Television and Radio Broadcasting of Ukraine dated 14.03.2024, is registered as a subject in the field of print media.
ID R30-03156

PUBLISHER

Liburkina L. M.

SITE SECTIONS

A Hybrid Approach to Modeling Cryptocurrency Market Dynamics Based on Cluster Analysis and Neural Network Technologies
Kochorba V. Y.

Kochorba, Valeriia Yu. (2026) “A Hybrid Approach to Modeling Cryptocurrency Market Dynamics Based on Cluster Analysis and Neural Network Technologies.” Business Inform 1:439–449.
https://doi.org/10.32983/2222-4459-2026-1-439-449

Section: Finance, Money Circulation and Credit

Article is written in Ukrainian
Downloads/views: 0

Download article (pdf) -

UDC 336.71:004.8

Abstract:
The aim of the article is to develop and practically test a set of models for analyzing the dynamics of the cryptocurrency market, based on a combination of data mining and deep learning methods. The relevance of the research is determined by the high volatility of crypto assets and the inefficiency of traditional econometric approaches in conditions of nonlinear market processes. The study employs descriptive statistics and correlation analysis methods to form the feature space; the k-means clustering method to classify cryptocurrencies by investment attractiveness; architectures of artificial neural networks (ANN) and long short-term memory (LSTM) networks to forecast time series of asset values. Data processing, model training, and result visualization were carried out using the Python programming language with the Pandas, Scikit-learn, Keras, and TensorFlow libraries. A clustering of the cryptocurrency market was performed based on capitalization, volatility, and historical profitability indicators. The optimal number of clusters was determined (k=5), and the quality of the clustering was confirmed using the silhouette coefficient (0.53). It was found that the most attractive assets for investment are those in cluster 5 (high profitability, moderate risk, representative – Axie) and cluster 3 (low risk, conservative strategy, representative – WBTC). Neural network models were created and trained for the representative coins of these clusters. It was found that the LSTM model demonstrates higher accuracy compared to the classical ANN, achieving a coefficient of determination R2>0,93 and a lower mean squared error (MSE) on the test data. The efficiency of applying «forget gates» in LSTM for filtering market noise and identifying long-term trends was demonstrated. The proposed hybrid approach enables the automation of the asset selection process for investment and enhances the accuracy of predicting their market value. The resulting findings can serve as a foundation for developing algorithmic trading strategies and decision support systems in investment portfolio management.

Keywords: financial market, cryptocurrency market, cluster analysis, k-means, neural networks, LSTM, forecasting, investment strategy.

Fig.: 10. Bibl.: 16.

Kochorba Valeriia Yu. – Candidate of Sciences (Economics), Associate Professor, Deputy Director, Educational and Scientific Institute «Karazin Banking Institute» of V. N. Karazin Kharkiv National University (55 Peremohy Ave., Kharkiv, 61174, Ukraine)
Email: [email protected]

List of references in article

Belcastro L., Carbone D., Cosentino C. & et al. (2023). Enhancing cryptocurrency price for ecasting by integrating machine learning with social media and market data. Algorithms, 12(16), Article 542. https://doi.org/10.3390/a16120542
Corbet S. (2019). Cryptocurrencies as a financial asset: A systematic analysis. International Review of Financial Analysis, 62, 182–199.
Dziundziuk B. V. (2023). Vykorystannia tekhnolohii blokchein yak informatsiinoi infrastruktury v publichnomu sektori [Use of blockchain technology as an information infrastructure in the public sector]. Derzhavne upravlinnia: udoskonalennia ta rozvytok, 4. http://doi.org/10.32702/2307-2156.2023.4.9
Hamayel M. J. & Owda A. Y. (2021). A novel cryptocurrency price prediction model using GRU, LSTM and bi LSTM machine learning algorithms. AI, 4(2), 477–496. https://doi.org/10.3390/ai2040030
Hansun S., Wicaksana A. & Khaliq A. Q. (2022). Multivariate cryptocurrency prediction: Comparative analysis of three recurrent neural networks approaches. Journal of Big Data, 9, 1–15.
Hitam N. A., Ismail A. R., Samsudin R. & Alkhammash E. H. (2022). The effect of kernel functions on cryptocurrency prediction using support vector machines // Proceedings of the International Conference of Reliable Information and Communication Technology. Springer, 319–332.
Kostenko O. V. & Radutnyi O. E. (2022). Blokchein i Metavsesvit: pravovi aspekty. [Blockchain and the Metaverse: Legal Aspects]. Yurydychnyi naukovyi elektronnyi zhurnal, 9, 499–506. https://doi.org/10.32782/2524-0374/2022-9/123
Lai S., Ye C. & Zhou H. J. H. (2021). Chinese stock trend prediction based on multi feature learning and model fusion // Proceedings of the 2021 IEEE International Conference on Smart Data Services (SMDS). Chicago, IL, USA, 18–23.
Murray K., Rossi A., Carraro D. & Visentin A. (2023). On forecasting cryptocurrency prices: a comparison of machine learning, deep learning, and ensembles. Forecasting, 1(5), 196–209. https://doi.org/10.3390/forecast5010010
Ozturk Birim S. (2022). An analysis for cryptocurrency price prediction using LSTM, GRU, and the bi directional implications // Developments in Financial and Economic Fields at the National and Global Scale. Gazi Kitabevi.
Payward, Inc. (2026). Kraken: ofitsiinyi sait kryptovaliutnoi birzhi [Kraken: official website of the cryptocurrency exchange]. Payward, Inc. https://www.kraken.com/uk
Regal A., Morzan J., Fabbri C. & et al. (2019). Cryptocurrency price projection based on tweets using LSTM. Ingeniare. Revista Chilena de Ingenieria, 4(27), 696–706. https://doi.org/10.4067/s0718-33052019000400696
Sadovyi R. Ya. (2024). Sutnist kryptovaliuty yak ekonomichnoi katehorii v mezhakh tekhnolohii blokchein [The Essence of Cryptocurrency as an Economic Category Within Blockchain Technology]. Scientific Bulletin of UNFU, 5(34), 52–61. https://doi.org/10.36930/40340507
Samsin R. I. (2023). Virtualni aktyvy: ryzyk vplyvu na finansovu stabilnist [Virtual Assets: Risk of Impact on Financial Stability]. Ekonomichna teoriia ta pravo, 4 (55), 56–72. https://doi.org/10.31359/2411-5584-2023-55-4-56
Seabe P. L., Moutsinga C. R. B. & Pindza E. (2023). Forecasting cryptocurrency prices using LSTM, GRU, and bi directional LSTM: a deep learning approach. Fractal and Fractional, 2(7), Article 203. https://doi.org/10.3390/fractalfract7020203
Singh A., Kumar A. & Akhtar Z. (2021). Bitcoin price prediction: A deep learning approach // Proceedings of the 2021 8th International Conference on Signal Processing and Integrated Networks (SPIN). Noida, India, 1053–1058.

FOR AUTHORS

License Contract

Conditions of Publication

Article Requirements

Regulations on Peer-Reviewing

Publication Contract

Current Issue

Frequently asked questions

INFORMATION

The Plan of Scientific Conferences

OUR PARTNERS

Journal «The Problems of Economy»

The site and its metadata are licensed under CC BY-SA.

Write to webmaster