Authors : Lawrence Chidiebere Nwodo; John Otozi Ugah; Stella Ebere Edeh; Ugboaja Samuel; Maduabuchi Ignatius Edeh

Volume/Issue : Volume 11 - 2026, Issue 9 - September

Google Scholar : https://tinyurl.com/26ks5xph

DOI : https://doi.org/10.38124/ijisrt/26sep754

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Cryptocurrency markets are characterized by high volatility, rapid price movements, fragmented market information, transaction costs, and strong sensitivity to investor sentiment, thereby making conventional trading approaches increasingly inadequate for autonomous decision making. This study developed and evaluated an artificial intelligence driven framework for autonomous cryptocurrency trading that integrates Deep Reinforcement Learning (DRL), sentiment analysis, and intelligent risk management within a unified trading environment. The framework employed a Deep QNetwork (DQN) to learn trading decisions from multiple market state variables, which included price movements, trading volume, volatility indicators, order-book information, and sentiment scores derived from external information sources. Unlike conventional trading systems that rely predominantly on historical numerical indicators, the proposed framework incorporated qualitative market intelligence from news, social media, community discussions, and web-based information to improve the contextual awareness of autonomous trading decisions. The system further integrates cost aware execution cost-aware execution mechanisms that account for transaction fees, slippage, market volatility, and execution latency, On the other hand intelligent risk controls such as stop-loss, take-profit, balance allocation, and trade-exposure limits are all incorporated to reduce potential losses and improve portfolio stability. The developed system was implemented as an autonomous web-based trading application and evaluated through controlled simulations using historical cryptocurrency data and simulated exchange environments. Results demonstrated that the DQN-based trading engine successfully generated autonomous buy, sell, and hold decisions and adapted to changing market conditions. The system also demonstrated improved execution efficiency through cost-minimization mechanisms, while risk-control features supported safer trading operations and portfolio protection.

Keywords : Artificial Intelligence; Cryptocurrency Trading; Deep Reinforcement Learning; Deep Q-Network; Sentiment Analysis; Risk Management in Cryptocurrency.

References :

1. F. Sabry et al., “Cryptocurrencies and artificial intelligence: Challenges and opportunities,” IEEE Access, vol. 8, pp. 175840–175858, 2020. doi: 10.1109/ACCESS.2020.3037437
2. Y. Sovbetov, “Factors influencing cryptocurrency prices: Evidence from Bitcoin, Ethrerum, Dash, Litecoin, and Monero,” Journal of Economics and Financial Analysis, vol.2, No 2, pp. 1-27, 2018. Doi:10.1991/jefa. v2i2.a16
3. J. Chu, S. Chan, S. Nadarajah, and J. Osterrieder “GARCH Modelling of Cryptocurrencies,” Journal of Risk and Financial Management, vol 10, No 17, pp.1-15, 2017. doi:10.3390/jrfm10040017
4. R. S. Sutton and A. G. Barto, Reinforcement Learning: An Introduction, 2nd ed. Cambridge, MA, USA: MIT Press, 2018.
5. V. Mnih et al., “Human-level control through deep reinforcement learning,” Nature, vol. 518, no. 7540, pp. 529–533, 2015. doi: 10.1038/nature14236
6. P. Weichbroth et al., “Security of cryptocurrencies: A view on the state-of-the-art research and current developments,” Sensors, vol. 23, no. 6, p. 3155, 2023. doi: 10.3390/s23063155
7. R. Hurjui, The Effect of Market Sentiment on Cryptocurrency Prices, Master’s thesis, Erasmus University Rotterdam, 2022. [Online]. Available: https://thesis.eur.nl/pub/61895/
8. G. Hu, “Advancing algorithmic trading: A multi-technique enhancement of deep Q network models,” arXiv preprint arXiv:2311.05743, Nov. 9, 2023. [Online]. Available: https://arxiv.org/abs/2311.05743
9. C. Liu, A. Mahanti, R. Naha, G. Wang, and E. Sbai, “Enhancing cryptocurrency sentiment analysis with multimodal features,” features,” arXiv preprint arXiv:2508.15825, 2025. [Online]. Available: https://arxiv.org/abs/2508.15825. doi: 10.48550/arXiv.2508.15825
10. M. B. Osman, C. Urom, K. Guesmi, and R. Benkraiem, “Economic sentiment and the cryptocurrency market in the post-COVID-19 era,” International Review of Financial Analysis, vol. 91, Art. no. 102962, 2024, doi: 10.1016/j.irfa.2023.102962.
11. F. Sabry, W. Labda, A. Erbad, and Q. Malluhi, “Cryptocurrencies and artificial intelligence: Challenges and opportunities,” IEEE Access, vol. 8, pp. 175840–175858, 2020, doi: 10.1109/ACCESS.2020.3025211.
12. Bashynska, M. Malanchuk, O. Zhuravel, and K. Olinichenko, “Smart solutions: Risk management of crypto-assets and blockchain technology,” International Journal of Civil Engineering and Technology (IJCIET), vol. 10, no. 2, pp. 1121–1131, Feb. 2019, Art. no. IJCIET_10_02_108. [Online]. Available: http://www.iaeme.com/ijciet/issues.asp?JType=IJCIET&VType=10&IType=2
13. O. Sattarov and J. Choi, “Multi-level deep Q-networks for Bitcoin trading strategies,” Scientific Reports, vol. 14, Art. no. 771, 2024, doi: 10.1038/s41598-024-51408-w.
14. Z. Jiang, D. Xu, and J. Liang, “A deep reinforcement learning framework for the financial portfolio management problem,” arXiv preprint arXiv: 1706.10059, 2017. doi: 10.48550/arXiv.1706.10059
15. L.C, Nwodo, J.O, Ugah, S.E, Edeh, and M.I, Edeh., “Design and implementation of AI-driven autonomous trading bots for cryptocurrency,” International Journal of Innovative Science and Research Technology, Vol 2, pp.583-588, 2026. https://doi.org/10.38124/ijsrt/26aug236
16. Organization for Economic Co-operation and Development, Explanatory memorandum on the updated definition of an AI system, OECD Artificial Intelligence Papers, no.8, Paris, France: OECD Publication, 2024.
17. S. Nakamoto, “Bitcoin: A peer-to-peer electronic cash system,” Bitcoin paper, 2008.
18. L.C. Nwodo, “Development of an autonomous trading web app for cryptocurrency markets,” Ph.D dissertation, Dept. Computer Science, Ebonyi State University, Abakaliki, Nigeria, August.2026.
19. S. Sun et al., “TradeMaster: A holistic quantitative trading platform empowered by reinforcement learning,” in Proc. 37th Conf. Neural Information Processing Systems (NeurIPS), 2023. [Online]. Available: https://proceedings.neurips.cc/paper_files/paper/2023/file/b8f6f7f2ba4137124ac976286eacb611-Paper-Datasets_and_Benchmarks.pdf
20. J.D. Gort, X. Y, Liu, J.Gao, S.Chen, and C.D, Wang, “ Deep reinforcement learning for cryptocurrency trading: Practical approach to address backtest overfitting,”, arXiv preprint, 2022.
21. C. Wade, “Using generative AI and deep learning for financial markets forecasting,” M.S. thesis, London School of Economics, 2022. [Online]. Available: https://etheses.lse.ac.uk/4657/1/Wade_200622083.pdf
22. Z. Jiang, D. Xu, and J. Liang, “A deep reinforcement learning framework for the financial portfolio management problem,” arXiv preprint arXiv: 1706.10059, 2017. doi: 10.48550/arXiv.1706.10059
23. X. H. Tay and S. M. Lim, “Deep reinforcement learning in cryptocurrency trading: A profitable approach,” J. Telecommunications Digital Economy, vol. 12, no. 3, pp.1-17, 2024.
24. S. Singh, “AI in financial trading: A review of recent advancements,” Int. J. Financial Technology, vol. 5, no. 2, pp. 134–150, 2023.