<link rel="stylesheet" href="styles.f3b1fba60ec7970c.css">

Сучасні підходи до інтелектуального аналізу ринку криптовалют

dc.contributor.authorDobroliubova, M. V.en
dc.contributor.authorRadovetskyi, O. O.en
dc.contributor.authorPomazun, O. M.en
dc.contributor.authorMarkin, M. O.en
dc.date.accessioned2026-04-14T13:18:45Z
dc.date.available2026-04-14T13:18:45Z
dc.date.issued2025
dc.description.abstractThe article addresses the challenges of intelligent analysis in the cryptocurrency market and, due to the limitations of traditional financial analysis methods, substantiates the need for the implementation of artificial intelligence (AI) systems. These systems are essential for processing large volumes of heterogeneous data, improving forecasting accuracy, and enhancing decision-making efficiency, particularly in light of the market’s high volatility, dynamic nature, and complexity. The review of recent studies on AI-based models for cryptocurrency price prediction reveals a growing demand for the development of flexible, hybrid architectures that integrate intelligent approaches with classical analytical techniques. The paper analyses the architecture of a hybrid AI system, encompassing modules for data collection, pre-processing, model training, and results visualization. It details the handling of various data types, including time series, on-chain metrics, and textual information. Within the framework of the empirical study, a comparative analysis of three models — ARIMA, Random Forest Regressor, and LSTM — was conducted using an open dataset of historical Bitcoin prices. Special attention is given to feature engineering, with the introduction of additional parameters that significantly improved model training performance. Based on the research findings, a concept of a hybrid multi-layered AI system is proposed. This system includes data acquisition and normalization modules, a computation block utilizing GPU/TPU resources for model training, and dashboards and APIs for result visualization and integration with exchange platforms. The proposed conceptual AI model can serve as a foundation for practical implementation in investor-oriented and financial-analytical platforms. It may attract the interest of fintech developers, traders, and researchers in the digital economy, paving the way for high-precision forecasting tools and automated decision-making solutions in the cryptocurrency marketen
dc.identifier.citationДобролюбова М. В., Радовецький О. О., Помазун О. М., Маркін М. О. Сучасні підходи до інтелектуального аналізу ринку криптовалют // Вісник Вінницького політехнічного інституту. 2025. № 4. С. 126-135. URI: https://visnyk.vntu.edu.ua/index.php/visnyk/article/view/3311.uk
dc.identifier.doihttps://doi.org/10.31649/1997-9266-2025-181-4-126-135
dc.identifier.issn1997-9274
dc.identifier.udc004.032.26
dc.identifier.urihttps://ir.lib.vntu.edu.ua/handle/123456789/51152
dc.language.isouk_UAuk_UA
dc.publisherВінницький національний технічний університетuk
dc.relation.ispartofВісник Вінницького політехнічного інституту. № 4 : 126-135.uk
dc.relation.referencesAna Todorovska, et al., “Using ML and Explainable AI to understand the interdependency networks between classical economic indicators and crypto-markets,” Physica A: Statistical Mechanics and its Applications, vol. 624, article 128900, Au-gust. 2023. https://doi.org/10.1016/j.physa.2023.128900.en
dc.relation.referencesA. Stencel, What is a meme coin? Dogecoin to the moon! , 2023. [Electronic resource]. Available: https://hal.science/hal-04360574v1/document. Accessed: March 17, 2025.en
dc.relation.referencesWaichung Chung, et al., “Ascertaining price formation in cryptocurrency markets with machine learning”, The European Journal of Finance, vol. 30, рр. 78-100, 2024. https://doi.org/10.1080/1351847X.2021.1908390.en
dc.relation.referencesK. Muneer, and U. Fatima, “Cryptocurrencies Analytics with Machine Learning and Human-centered Explainable AI: Enhancing Decision-Making in Dynamic Market”, International Journal of Computer Applications, vol. 186, no. 62, pp. 52-67, 2025. https://doi.org/10.5120/ijca2025924418.en
dc.relation.referencesM. Zubair, et al., “An Improved Machine Learning-Driven Framework for Cryptocurrencies Price Prediction with Senti-mental Cautioning”, IEEE Access, vol. 12, pp. 51395-51418, 2024. https://doi.org/10.1109/access.2024.3367129.en
dc.relation.referencesNrusingha Tripathy, et al., “Cryptocurrency Price Prediction: A Comprehensive Analysis of High-dimensional Features with Deep Learning Techniques,” in 2024 International Conference on Intelligent Computing and Sustainable Innovations in Technology (IC-SIT). Conference Proceedings, 2024, pp. 1-6. http://doi.org/10.1109/IC-SIT63503.2024.10862895.en
dc.relation.referencesR. Amirzadeh, A. Nazari, and D. Thiruvady, “Applying Artificial Intelligence in Cryptocurrency Markets: a Sur-vey,”Algorithms, vol. 15, no. 11, рр. 1-26, 2022. https://doi.org/10.3390/a15110428.en
dc.relation.referencesJ. Fattah et al., “Forecasting of demand using ARIMA model,” International Journal of Engineering Business Manage-ment, vol. 10, рр. 1-9, 2018. https://doi.org/10.1177/1847979018808673.en
dc.relation.referencesY. Chen, and J. Li, “Recurrent Neural Networks algorithms and applications,” in 2nd International Conference on Big Data & Artificial Intelligence & Software Engineering (ICBASE). Conference proceedings, Zhuhai, China, 2021, pp. 38-43. http://doi.org/10.1109/ICBASE53849.2021.00015.en
dc.relation.referencesS. Hochreiter, and J. Schmidhuber, “Long Short-Term Memory,” Neural Computation, vol. 9, no. 8, рр. 1735-1780, 1997. https://doi.org/10.1162/neco.1997.9.8.1735.en
dc.relation.referencesL. Breiman, “Random Forests,” Machine Learning, vol. 45, no. 1, рр. 5-32, 2001. https://doi.org/10.1023/a:1010933404324.en
dc.relation.referencesXavier Amatriain, “Transformer models: an introduction and catalog,” arXiv (a curated research-sharing platform)2024, https://doi.org/10.48550/arXiv.2302.07730.en
dc.relation.referencesWhat are the settings in the Market Making bot| Cryptohopper Documentation. Cryptohopper Documentation. [Electronic resource]. Available: https://docs.cryptohopper.com/docs/market-making/what-are-the-settings-in-the-market-making-bot. Accessed: 30.04.2025.en
dc.relation.urihttps://visnyk.vntu.edu.ua/index.php/visnyk/article/view/3311
dc.subjectcryptocurrencyen
dc.subjectdataseten
dc.subjectmachine learningen
dc.subjectdeep learningen
dc.subjectneural networksen
dc.subjectAI systemsen
dc.subjectintelligent data analysisen
dc.subjectdatabaseen
dc.subjectprogrammingen
dc.titleСучасні підходи до інтелектуального аналізу ринку криптовалютuk
dc.title.alternativeModern Approaches to the Intelligent Analysis of the Cryptocurrency Marketen
dc.typeArticle, professional native edition
dc.typeArticle

Файли

Контейнер файлів

Зараз показуємо 1 - 1 з 1
Вантажиться...
Ескіз
Назва:
191677.pdf
Розмір:
836,96 KB
Формат:
Adobe Portable Document Format

Ліцензійна угода

Зараз показуємо 1 - 1 з 1
Вантажиться...
Ескіз
Назва:
license.txt
Розмір:
129 B
Формат:
Plain Text
Опис: