<i>F-LSTM</i>: Federated learning-based LSTM framework for cryptocurrency price prediction

Nihar Patel; Nakul Vasani; Nilesh Kumar Jadav; Rajesh Gupta; Sudeep Tanwar; Zdzislaw Polkowski; Fayez Alqahtani; Amr Gafar; Nihar Patel; Nakul Vasani; Nilesh Kumar Jadav; Rajesh Gupta; Sudeep Tanwar; Zdzislaw Polkowski; Fayez Alqahtani; Amr Gafar

doi:10.3934/era.2023330

Electronic Research Archive

2023, Volume 31, Issue 10: 6525-6551. doi: 10.3934/era.2023330

Previous Article Next Article

Research article

F-LSTM: Federated learning-based LSTM framework for cryptocurrency price prediction

1.
Department of CSE, Institute of Technology, Nirma University, Ahmedabad, India
2.
Department of Humanities and Social Sciences, The Karkonosze University of Applied Sciences in Jelenia GÅLora, Poland
3.
Software Engineering Department, College of Computer and Information Sciences, King Saud University, Riyadh 11437, Saudi Arabia
4.
Mathematics and Computer Science Department, Faculty of Science, Menofia University, Shebin Elkom, Egypt

Received: 09 July 2023 Revised: 21 September 2023 Accepted: 26 September 2023 Published: 11 October 2023

In this paper, a distributed machine-learning strategy, i.e., federated learning (FL), is used to enable the artificial intelligence (AI) model to be trained on dispersed data sources. The paper is specifically meant to forecast cryptocurrency prices, where a long short-term memory (LSTM)-based FL network is used. The proposed framework, i.e., F-LSTM utilizes FL, due to which different devices are trained on distributed databases that protect the user privacy. Sensitive data is protected by staying private and secure by sharing only model parameters (weights) with the central server. To assess the effectiveness of F-LSTM, we ran different empirical simulations. Our findings demonstrate that F-LSTM outperforms conventional approaches and machine learning techniques by achieving a loss minimal of $ 2.3 \times 10^{-4} $. Furthermore, the F-LSTM uses substantially less memory and roughly half the CPU compared to a solely centralized approach. In comparison to a centralized model, the F-LSTM requires significantly less time for training and computing. The use of both FL and LSTM networks is responsible for the higher performance of our suggested model (F-LSTM). In terms of data privacy and accuracy, F-LSTM addresses the shortcomings of conventional approaches and machine learning models, and it has the potential to transform the field of cryptocurrency price prediction.
- federated learning,
- LSTM,
- decentralization,
- cryptocurrencies,
- weight sharing
Citation: Nihar Patel, Nakul Vasani, Nilesh Kumar Jadav, Rajesh Gupta, Sudeep Tanwar, Zdzislaw Polkowski, Fayez Alqahtani, Amr Gafar. F-LSTM: Federated learning-based LSTM framework for cryptocurrency price prediction[J]. Electronic Research Archive, 2023, 31(10): 6525-6551. doi: 10.3934/era.2023330

Related Papers:

Abstract

In this paper, a distributed machine-learning strategy, i.e., federated learning (FL), is used to enable the artificial intelligence (AI) model to be trained on dispersed data sources. The paper is specifically meant to forecast cryptocurrency prices, where a long short-term memory (LSTM)-based FL network is used. The proposed framework, i.e., F-LSTM utilizes FL, due to which different devices are trained on distributed databases that protect the user privacy. Sensitive data is protected by staying private and secure by sharing only model parameters (weights) with the central server. To assess the effectiveness of F-LSTM, we ran different empirical simulations. Our findings demonstrate that F-LSTM outperforms conventional approaches and machine learning techniques by achieving a loss minimal of $ 2.3 \times 10^{-4} $. Furthermore, the F-LSTM uses substantially less memory and roughly half the CPU compared to a solely centralized approach. In comparison to a centralized model, the F-LSTM requires significantly less time for training and computing. The use of both FL and LSTM networks is responsible for the higher performance of our suggested model (F-LSTM). In terms of data privacy and accuracy, F-LSTM addresses the shortcomings of conventional approaches and machine learning models, and it has the potential to transform the field of cryptocurrency price prediction.

References

[1]	K. Pilbeam, Finance and Financial Markets, Bloomsbury Publishing, 2018.
[2]	M. Watorek, S. Drozdz, J. Kwapien, L. Minati, P. Oswiecimka, M. Stanuszek, Multiscale characteristics of the emerging global cryptocurrency market, Phys. Rep., 901 (2021), 1–82. https://doi.org/10.1016/j.physrep.2020.10.005 doi: 10.1016/j.physrep.2020.10.005
[3]	M. Resta, P. Pagnottoni, M. E. D. Giuli, Technical analysis on the bitcoin market: Trading opportunities or investors' pitfall, Risks, 8 (2020), 44. https://doi.org/10.3390/risks8020044 doi: 10.3390/risks8020044
[4]	N. Miller, Y. Yang, B. Sun, G. Zhang, Identification of technical analysis patterns with smoothing splines for bitcoin prices, J. Appl. Stat., 46 (2019), 2289–2297. https://doi.org/10.1080/02664763.2019.1580251 doi: 10.1080/02664763.2019.1580251
[5]	R. Parekh, N. P. Patel, N. Thakkar, R. Gupta, S. Tanwar, G. Sharma, et al., DL-Guess: Deep learning and sentiment analysis-based cryptocurrency price prediction, IEEE Access, 10 (2022), 35398–35409. https://doi.org/10.1109/ACCESS.2022.3163305 doi: 10.1109/ACCESS.2022.3163305
[6]	B. M. Henrique, V. A. Sobreiro, H. Kimura, Literature review: Machine learning techniques applied to financial market prediction, Expert Syst. Appl., 124 (2019), 226–251. https://doi.org/10.1016/j.eswa.2019.01.012 doi: 10.1016/j.eswa.2019.01.012
[7]	L. Alessandretti, A. ElBahrawy, L. M. Aiello, A. Baronchelli, Anticipating cryptocurrency prices using machine learning, Complexity, 2018 (2018), 8983590. https://doi.org/10.1155/2018/8983590 doi: 10.1155/2018/8983590
[8]	T. E. Koker, D. Koutmos, Cryptocurrency trading using machine learning, J. Risk Financ. Manage., 13 (2020), 178. https://doi.org/10.3390/jrfm13080178 doi: 10.3390/jrfm13080178
[9]	N. P. Patel, R. Parekh, N. Thakkar, R. Gupta, S. Tanwar, G. Sharma, et al., Fusion in cryptocurrency price prediction: A decade survey on recent advancements, architecture, and potential future directions, IEEE Access, 10 (2022), 34511–34538. https://doi.org/10.1109/ACCESS.2022.3163023 doi: 10.1109/ACCESS.2022.3163023
[10]	D. H. Kwon, J. B. Kim, J. S. Heo, C. M. Kim, Y. H. Han, Time series classification of cryptocurrency price trend based on a recurrent LSTM neural network, J. Inf. Process. Syst., 15 (2019), 694–706. https://doi.org/10.3745/JIPS.03.0120 doi: 10.3745/JIPS.03.0120
[11]	E. S. Pour, H. Jafari, A. Lashgari, E. Rabiee, A. Ahmadisharaf, Cryptocurrency price prediction with neural networks of LSTM and bayesian optimization, Eur. J. Bus. Manage. Res., 7 (2022), 20–27. https://doi.org/10.24018/ejbmr.2022.7.2.1307 doi: 10.24018/ejbmr.2022.7.2.1307
[12]	S. P. Ardakani, N. Du, C. Lin, J. C. Yang, Z. Bi, L. Chen, A federated learning-enabled predictive analysis to forecast stock market trends, J. Ambient Intell. Hum. Comput., 14 (2023), 1–7. https://doi.org/10.1007/s12652-023-04570-4 doi: 10.1007/s12652-023-04570-4
[13]	W. Jiang, Applications of deep learning in stock market prediction: Recent progress, Expert Syst. Appl., 184 (2021), 115537. https://doi.org/10.1016/j.eswa.2021.115537 doi: 10.1016/j.eswa.2021.115537
[14]	W. An, L. Wang, D. Zhang, Comprehensive commodity price forecasting framework using text mining methods, J. Forecasting, 42 (2023), 1865–1888. https://doi.org/10.1002/for.2985 doi: 10.1002/for.2985
[15]	Q. Zhang, C. Qin, Y. Zhang, F. Bao, C. Zhang, P. Liu, Transformer-based attention network for stock movement prediction, Expert Syst. Appl., 202 (2022), 117239. https://doi.org/10.1016/j.eswa.2022.117239 doi: 10.1016/j.eswa.2022.117239
[16]	S. Halder, FinBERT-LSTM: Deep learning based stock price prediction using news sentiment analysis, arXiv preprint, (2022), arXiv: 2211.07392. https://doi.org/10.48550/arXiv.2211.07392
[17]	Z. Shi, Y. Hu, G. Mo, J. Wu, Attention-based CNN-LSTM and XGBoost hybrid model for stock prediction, arXiv preprint, (2023), arXiv: 2204.02623. https://doi.org/10.48550/arXiv.2204.02623
[18]	P. Nithyakani, R. J. Tom, P. Gupta, A. Shanthini, V. M. John, V. Sharma, Prediction of bitcoin price using Bi-LSTM network, in 2021 International Conference on Computer Communication and Informatics (ICCCI), Coimbatore, India, (2021), 1–5. https://doi.org/10.1109/ICCCI50826.2021.9402427
[19]	P. Bhattacharya, S. B. Patel, R. Gupta, S. Tanwar, J. J. P. C. Rodrigues, SATYA: Trusted bi-lstm-based fake news classification scheme for smart community, IEEE Trans. Comput. Social Syst., 9 (2022), 1758–1767. https://doi.org/10.1109/TCSS.2021.3131945 doi: 10.1109/TCSS.2021.3131945
[20]	C. Luo, L. Pan, B. Chen, H. Xu, Bitcoin price forecasting: An integrated approach using hybrid LSTM-ELM models, Math. Probl. Eng., 2022 (2022), 2126518. https://doi.org/10.1155/2022/2126518 doi: 10.1155/2022/2126518
[21]	I. E. Livieris, N. Kiriakidou, S. Stavroyiannis, P. Pintelas, An advanced CNN-LSTM model for cryptocurrency forecasting, Electronics, 10 (2021), 287. https://doi.org/10.3390/electronics10030287 doi: 10.3390/electronics10030287
[22]	F. Ferdiansyah, S. H. Othman, R. Zahilah Raja Md Radzi, D. Stiawan, Y. Sazaki, U. Ependi, A LSTM-method for bitcoin price prediction: A case study yahoo finance stock market, in 2019 International Conference on Electrical Engineering and Computer Science (ICECOS), (2019), 206–210. https://doi.org/10.1109/ICECOS47637.2019.8984499
[23]	Pip install yfinance, YFinance 0.2.21, pip install yfinance, Available from: https://pypi.org/project/yfinance/.
[24]	L. Peng, L. Wang, D. Xia, Q. Gao, Effective energy consumption forecasting using empirical wavelet transform and long short-term memory, Energy, 238 (2022), 121756. https://doi.org/10.1016/j.energy.2021.121756 doi: 10.1016/j.energy.2021.121756

Reader Comments

Your name:*

Email:*
© 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)