کاهش درصد خطای پیشبینی سریهای زمانی قیمت رمزارزها با استفاده از دوسویهسازی شبکههای عصبی یادگیری عمیق
محورهای موضوعی : electrical and computer engineeringفتانه کاظم زاده 1 , مسعود هوشمند کفاشیان 2 , منیره هوشمند 3 *
1 - دانشگاه بین المللی امام رضا (ع)
2 - شرکت مخابرات، مشهد
3 - دانشگاه بین المللی امام رضا (ع)
کلید واژه: پیشبینی سریهای زمانی, یادگیری عمیق, شبکه عصبی دوطرفه, پیشبینی قیمت رمزارزها, خطای پیشبینی شبکه عصبی, درصد خطای پیشبینی قیمت.,
چکیده مقاله :
پیشبینی سریهای زمانی در حوزههای مهندسی، مخابرات و امور مالی از اهمیت بالایی برخوردار است. سریهای زمانی مالی، که اغلب چند متغیره هستند، نیاز به الگوریتمهای دقیق و بهینه دارند. در پژوهشهای سالهای اخیر، شبکههای عصبی عمیق در بهبود دقت پیشبینی سریهای زمانی مالی نتایج موفقی نشان دادهاند. این پژوهش به بررسی استفاده از شبکههای LSTM و GRU در پیشبینی قیمت رمزارزها پرداخته و رویکرد دوسویهسازی این شبکهها را با تاکید به انتخاب بهینه هایپرپارامترها به منظور کاهش خطای پیشبینی و افزایش دقت با استفاده از روشهای جستجوی شبکهای، جستجوی اتفاقیCV و بیزین مورد مطالعه قرار میدهد. نتایج شبیهسازی نشان میدهند که استفاده از شبکههای دو سویه LSTM و شبکه دوسویه BiGRU کاهش درصد خطا را برای رمزارز BTC تا 22/3% ، برای ETH تا 94/3% ، و برای LTC تا 99/3% به همراه داشته است.
Time series forecasting in engineering, telecommunications, and finance is of great importance. Financial time series, which are often multivariate, require precise and optimized algorithms. In recent research, deep neural networks have demonstrated successful results in improving the accuracy of financial time series forecasting. This study investigates the use of LSTM and GRU networks in predicting cryptocurrency prices and examines the bidirectional implementation of these networks, with an emphasis on optimal hyperparameter selection to reduce prediction error and enhance accuracy through Grid Search, RandomizedSearchCV, and Bayesian methods. Simulation results indicate that employing bidirectional LSTM and BiGRU networks reduced the prediction error rate for BTC by up to 3.22%, for ETH by 3.94%, and for LTC by 3.99%.
[1] A. Gasparin, S. Lukovic, and C. Alippi, "Deep learning for time series forecasting: The electric load case," CAAI Trans. on Intelligence Technology, vol. 7, no. 1, pp. 1-25, Mar. 2022.
[2] M. Shin, D. Mohaisen, and J. Kim, "Bitcoin price forecasting via ensemble-based LSTM deep learning networks," in Proc. Int. Conf. on Information Networking, pp. 603-608, Jeju Island, South Korea, 13-16 Jan. 2021.
[3] S. Yang, X. Yu, and Y. Zhou, "LSTM and GRU neural network performance comparison study: Taking Yelp review dataset as an example," in Proc. Int. Workshop on Electronic Communication and Artificial Intelligence, pp. 98-101, Shanghai, China, 12-14 Jun. 2020.
[4] T. Shintate and L. Pichl, "Trend prediction classification for high frequency bitcoin time series with deep learning," Journal of Risk and Financial Management, vol. 12, no. 1, Article ID: 17, Mar. 2019.
[5] K. A. Althelaya, E.-S. M. El-Alfy, and S. Mohammed, "Stock market forecast using multivariate analysis with bidirectional and stacked (LSTM, GRU)," in Proc. 21st Saudi Computer Society National Computer Conf., 7 pp., Riyadh, Saudi Arabia, 25-26 Apr. 2018.
[6] A. Singh, A. Kumar, and Z. Akhtar, "Bitcoin price prediction: A deep learning approach," in Proc. 8th Int. Conf. on Signal Processing and Integrated Networks, pp. 1053-1058, Noida, India, 26-27 Aug. 2021.
[7] G. Liu, F. Xiao, C. –T. Lin, and Z. Cao, "A fuzzy interval time-series energy and financial forecasting model using network-based multiple time-frequency spaces and the induced-ordered weighted averaging aggregation operation," IEEE Trans. on Fuzzy Systems, vol. 28, no. 11, pp. 2677-2690, Nov. 2020.
[8] R. Reyhani and A. M. E. Moghadam, "A heuristic method for forecasting chaotic time series based on economic variables," in Proc. 6th Int. Conf. on Digital Information Management, pp. 300-304, Melbourne, Australia, 26-28 Sept. 2011.
[9] W. He, et al., "Applying multiple time series data mining to large-scale network traffic analysis," in Proc. IEEE Conf. on Cybernetics and Intelligent Systems, pp. 394-399, Chengdu, China, 21-24 Sept.2008.
[10] J. Bergstra and Y. Bengio, "Random search for hyper-parameter optimization," Journal of Machine Learning Research, vol. 13, pp. 281-305, 2012.
[11] J. Snoek, H. Larochelle, and R. P. Adams, "Practical Bayesian optimization of machine learning algorithms," in Proc. 26th Int. Conf. on Neural Information Processing Systems, vol. 2, pp. 2951- 2959, Lake Tahoe, NV, USA, 3-6 Dec. 2012.
[12] P. L. Seabe, C. R. B. Moutsinga, and E. Pindza, "Forecasting cryptocurrency prices using LSTM, GRU, and bi-directional LSTM: a deep learning approach," Fractal and Fractional, vol. 7, no. 2, Article ID: 203, Feb. 2023.
[13] M. J. Hamayel and A. Y. Owda, "A novel cryptocurrency price prediction model using GRU, LSTM and bi-LSTM machine learning algorithms," AI, vol. 2, no. 4, pp. 477-496, Dec. 2021. [14] M. Rafi, et al., "Enhancing cryptocurrency price forecasting accuracy: A feature selection and weighting approach with bi-directional LSTM and trend-preserving model bias correction," IEEE Access, vol. 11, pp. 65700-65710, 2023.
[15] N. Hussein and A. M. Abdulazeez, "Bitcoin price prediction using hybrid LSTM-GRU models," The Indonesian Journal of Computer Science, vol. 13, no. 1, pp. 94-101, 2024.
[16] M. H. A. Hadi, N. A. Ramli, and Q. U. Islam, "Predicting Bitcoin and Ethereum prices using long short-term memory and gated recurrent unit," Data Analytics and Applied Mathematics, vol. 4, no. 2, pp. 8-17, 2023.
[17] P. Pandey and G. Sharma, "Effective price prediction of cryptocurrencies using CNN-based dual directional model," Science & Technology Asia, vol. 30, no. 1, pp. 201-219, Jan./Mar. 2025.
[18] X. Wang, I. Cretu, and H. Meng, "A cryptocurrency price forecasting model by integrating empirical mode decomposition and LSTM neural networks," Artificial Intelligence and Applications, vol. 3, no. 3, pp. 305-315, Mar. 2025.
[19] S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural Computation, vol. 9, no. 8, pp. 1735-1780, 1997.
[20] K. Cho, et al., Learning Phrase Representations Using RNN Encoder-Decoder for Statistical Machine Translation, arXiv preprint arXiv:1406.1078, 2014.
[21] W. Zheng and G. Chen, "An accurate GRU-based power time-series prediction approach with selective state updating and stochastic optimization," IEEE Trans. on Cybernetics, vol. 52, no. 12, pp. 13902-13914, Dec. 2022.
[22] S. Hansun, A. Wicaksana, and A. Q. Khaliq, "Multivariate cryptocurrency prediction: comparative analysis of three recurrent neural networks approaches," Journal of Big Data, vol. 9, no. 1, Article ID: 50, 2022.
[23] S. Ozturk Birim, "An analysis for cryptocurrency price prediction using LSTM, GRU, and the bi-directional implications," In S. Karabulut (ed.), Developments in Financial and Economic Fields at the National and Global Scale, pp. 377-392, 2022.
[24] T. Fischer and C. Krauss, "Deep learning with long short-term memory networks for financial market predictions," European Journal of Operational Research, vol. 270, no. 2, pp. 654-669, Oct. 2018.
[25] S. Siami-Namini, N. Tavakoli, and A. S. Namin, "A comparison of ARIMA and LSTM in forecasting time series," in Proc. 17th IEEE Int. Conf. on Machine Learning and Applications, pp. 1394-1401, Orlando, FL, USA, 17-20 Dec. 2018.
[26] S. McNally, J. Roche, and S. Caton, "Predicting the price of bitcoin using machine learning," in 2018 26th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), 2018.
[27] D. Bahdanau, K. Cho, and Y. Bengio, Neural Machine Translation by Jointly Learning to Align and Translate, arXiv preprint arXiv:1409.0473, 2014.