ارائه یک مدل ترکیبی غیرقطعی برای پیشبینی قیمت رمز ارز بیتکوین CNN-LSTM
محورهای موضوعی : مدیریت
علی علی جماعت
1
(نویسنده مسئول، استادیار گروه کامپیوتر، دانشکده فنی و مهندسی، واحد ابهر، دانشگاه آزاد اسلامی، ابهر ، ایران. آدرس پست
jamaat@kiau.ac.ir : الکترونیک)
سید محسن میرحسینی
2
(استادیار گروه کامپیوتر، دانشکده فنی و مهندسی، واحد هیدج، دانشگاه آزاد اسلامی، هیدج، ایران.)
کلید واژه: کلمات کلیدی: پیشبینی قیمت, بیتکوین, یادگیری ماشین و یادگیری عمیق,
چکیده مقاله :
چکیدهدر جامع ه امروزی تنوع سرما یهگذا ری اهمیت بالایی یافته است . افرا د با تنو عبخشی به سبد سرمایه، ریسکسرما یهگذاری را کاهش م یدهند. بی تکوین نیز ب هعنوان یکی از سرما یههای دیجیتالی محبوبیت زیادی به دستآورده و در سبد سرمای هگذاری افراد و نهادها قرارگرفته است. پی شبینی قیمت بیتکوین برای تعیین روندقیمتی و معاملات مهم است. برای ا ین کار رو شهای سنتی و نیز روشهای مبتنی بر یادگیری ماشین مختلفیارائه شده است که هرکدام مزایا و معا یب خود را دارن د. اخیرا استفاده از مد لهای ترکیبی مورد توجه قرارگرفته است. روشهای ترکیبی کارایی مناسبی داشته و از مزایای روشهای ترکیب شده استفاده میکنند. دراین مقاله، یک روش ترکیبی مبتنی بر شبکه عصبی عمیق کانولوشنی و شبکه عصبی بازگشتی با حذف تصادفیاحتمالی ارائه م یشود. حذف تصادفی احتمالی موجب منظمسازی یادگیری و پرهیز از بیش برازش شده وموجب کاهش خطای مدل م یشود. نتایج آزمایشهای انجام گرفته نشاندهنده دقت بالاتر روش پیشنهادینسبت به رو شهای مورد مقایسه در پیشبینی قیمت بیت کوین دارد .کلمات کلیدی: پیشبینی قیمت، بیتکوین، یادگیری ماشین و یادگیری عمیق .
AbstractIn today's society, investment diversity has become very important. People reduceinvestment risk by diversifying their portfolios. Bitcoin has gained muchpopularity as one of the digital capitals and has been included in the investmentportfolio of individuals and institutions. Bitcoin price prediction is essential fordetermining price trends and transactions. For this purpose, various traditionalmethods as well as methods based on machine learning have been presented, eachof which has its own advantages and disadvantages. Recently, the use of hybridmodels has received attention. Combined methods have good efficiency and usethe advantages of combined techniques. This paper presents a hybrid methodbased on a deep convolutional neural network and recurrent neural network withprobabilistic dropout. Eliminating possible randomness leads to the regularizationof learning, avoids overfitting, and reduces model error. The results of theexperiments show that the proposed method has a higher accuracy than thecompared methods in predicting the price of Bitcoin.
منابع
-وزان، م یلاد( 1400 ) . یادگیری ماشین و علم داده: مبانی، مفاهیم، الگور یتمها و ابزاره، تهران: میعاد اندیشه،
.1400
- Alarab, I., Prakoonwit, S., Nacer, M.I. (2021). Illustrative Discussion of MC-Dropout
in General Dataset: Uncertainty Estimation in Bitcoin. Neural Process Lett 53, 1001–
1011, https://doi.org/10.1007/s11063-021-10424-x
- Albariqi, R., Winarko, E., (2020, February). “Prediction of Bitcoin Price Change using
Neural Networks”, In 2020 International Conference on Smart Technology and
Applications (ICoSTA) (pp. 1-4). IEEE.
- Bonneau, J., Miller, A., Clark, J., Narayanan, A., Kroll, J. A., Felten, E. W. (2015, May).
Sok: Research perspectives and challenges for bitcoin and cryptocurrencies, In 2015
IEEE Symposium on Security and Privacy (pp. 104-121). IEEE.
- Chung J., Gulcehre, C., Cho, K., Bengio Y.,"Empirical evaluation of gated recurrent
neural networks on sequence modeling," arXiv preprint arXiv:1412.3555, 2014.
- Chollet, F. (2021). Deep Learning with Python. Manning Publications Co., .202118-
- Joshi, R, Accuracy, precision (2016). recall & f1 score: Interpretation of performance
measures, Retrieved April.
- Hochreiter, S., Schmidhuber, J., (1997). "Long short-term memory," Neural
computation, Vol. 9, No. 8, pp.1735-1780.
- Kim, P. (2021).Neural Network, MATLAB Deep Learning, pp. 19-51, 2017.
- Koo, E., Kim, G. (202). A Hybrid Prediction Model Integrating GARCH Models with
a Distribution Manipulation Strategy Based on LSTM Networks for Stock Market
Volatility. IEEE Access, 10, 34743–34754. [CrossRef] ÷
- Kourentzes, N., Barrow, D.K., Crone, S,F., (2014). Neural network ensemble operators
for time series forecasting, Expert Systems with Applications, Vol. 41, No. 9, 4235-4244.
doi:10.1016/j.eswa.2013.12.011
- Lemay, A., Hoebel, K., Bridge, C.P. et al. Improving the repeatability of deep learning
models with Monte Carlo dropout. npj Digit. Med. 5, 174 (2022).
https://doi.org/10.1038/s41746-022-00709-3
- Livieris, I.E.,Kiriakidou, N., (2021).An advanced CNN-LSTM model for
cryptocurrency forecasting. Electronics 2021, 10, 287.
- Liu, L., Gong, C., L. Yang, and Y. Chen, (2020). DSTP-RNN: A dual-stage two-phase
attention-based recurrent neural network for long-term and multivariate time series
prediction, Expert Systems with Applications,Vol. 143, p. 113082
- Lunesu and M. Marchesi (2015). Bitcoin Spread Prediction Using Social and Web
Search Media, in UMAP Workshops
- Madan, I., Saluja, S., A. Zhao A. (2015). Automated bitcoin trading via machine learning
algorithms, URL. Matta, M, I.
- McNally, S., Roche, J., Caton, S. (2018, March). “Predicting the price of bitcoin using
machine learning”, In 2018 26th Euromicro International Conference on Parallel,
Distributed and Network-based Processing (PDP) (pp. 339-343). IEEE.
غیرقطعی برای پی شبینی قیم ت.. . CNN-LSTM 176 / ............ . ارائه یک مدل ترکیبی
- Mittal, M.,Geetha, G., (2022).Predicting Bitcoin Price using Machine Learning, 2022
International Conference on Computer Communication and Informatics (ICCCI), pp. 1-
7, doi: 10.1109/ICCCI54379.2022.9740772.
- Gal, Y., Ghahramani Z., (2016).Dropout as a bayesian approximation: representing
model uncertainty in deep learning. In: international conference on machine learning, pp
1050–1059
- Greaves, A., Au, B. (2015). Using the bitcoin transaction graph to predict the price of
bitcoin, No Data.
- Patel, M.; Tanwar, S. (2020). A deep learning-based cryptocurrency price prediction
scheme for financial institutions. J. Inf. Secur., 55, 102583.
- Peng, Y., Henrique, P., Albuquerque, M.,,(2018).The best of two worlds: Forecasting
high frequency volatility for cryptocurrencies and traditional currencies with Support
Vector Regression, Expert Systems with Applications, Vol. 97, 177-192. doi:
10.1016/j.eswa.2017.12.004
- Sharifi, A. M., Khalili Damghani, K., Abdi, F., Sardar, S. (2022). 'A hybrid model for
predicting bitcoin price using machine learning and metaheuristic algorithms', Journal of
Applied Research on Industrial Engineering, 9(1), pp. 134-150. doi:
10.22105/jarie.2021.291175.1343
- Sin, E., Wang, L., (2017, July). “Bitcoin price prediction using ensembles of neural
networks”, In 2017 13th International Conference on Natural Computation, Fuzzy
Systems and Knowledge Discovery (ICNC-FSKD) (pp. 666-671). IEEE
- Srivastava, N., (2013). Improving Neural Networks with Dropout, Master, Toronto,
Canada, 2013.
- Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.
(2014).Dropout: a simple way to prevent neural networks from overfitting, J Mach Learn
Res 15(1):1929–1958
- Sovbetov, Y. (2018). Factors influencing cryptocurrency prices: Evidence from bitcoin,
ethereum, dash, litcoin, and monero, Journal of Economics and Financial Analysis, 2(2),
1-27.
- Xiong, L., Lu, Y. (2017, April). “Hybrid ARIMA-BPNN model for time series
prediction of the Chinese stock market”, In 2017 3rd International Conference on
Information Management (ICIM) (pp. 93-97). IEEE
- Alarab, I., Prakoonwit, S., Nacer, M.I. (2021). Illustrative Discussion of MC-Dropout
in General Dataset: Uncertainty Estimation in Bitcoin. Neural Process Lett 53, 1001–
1011, https://doi.org/10.1007/s11063-021-10424-x
- Albariqi, R., Winarko, E., (2020, February). “Prediction of Bitcoin Price Change using
Neural Networks”, In 2020 International Conference on Smart Technology and
Applications (ICoSTA) (pp. 1-4). IEEE.
- Bonneau, J., Miller, A., Clark, J., Narayanan, A., Kroll, J. A., Felten, E. W. (2015, May).
Sok: Research perspectives and challenges for bitcoin and cryptocurrencies, In 2015
IEEE Symposium on Security and Privacy (pp. 104-121). IEEE.
- Chung J., Gulcehre, C., Cho, K., Bengio Y.,"Empirical evaluation of gated recurrent
neural networks on sequence modeling," arXiv preprint arXiv:1412.3555, 2014.
- Chollet, F. (2021). Deep Learning with Python. Manning Publications Co., .202118-
- Joshi, R, Accuracy, precision (2016). recall & f1 score: Interpretation of performance
measures, Retrieved April.
- Hochreiter, S., Schmidhuber, J., (1997). "Long short-term memory," Neural
computation, Vol. 9, No. 8, pp.1735-1780.
- Kim, P. (2021).Neural Network, MATLAB Deep Learning, pp. 19-51, 2017.
- Koo, E., Kim, G. (202). A Hybrid Prediction Model Integrating GARCH Models with
a Distribution Manipulation Strategy Based on LSTM Networks for Stock Market
Volatility. IEEE Access, 10, 34743–34754. [CrossRef] ÷
- Kourentzes, N., Barrow, D.K., Crone, S,F., (2014). Neural network ensemble operators
for time series forecasting, Expert Systems with Applications, Vol. 41, No. 9, 4235-4244.
doi:10.1016/j.eswa.2013.12.011
- Lemay, A., Hoebel, K., Bridge, C.P. et al. Improving the repeatability of deep learning
models with Monte Carlo dropout. npj Digit. Med. 5, 174 (2022).
https://doi.org/10.1038/s41746-022-00709-3
- Livieris, I.E.,Kiriakidou, N., (2021).An advanced CNN-LSTM model for
cryptocurrency forecasting. Electronics 2021, 10, 287.
- Liu, L., Gong, C., L. Yang, and Y. Chen, (2020). DSTP-RNN: A dual-stage two-phase
attention-based recurrent neural network for long-term and multivariate time series
prediction, Expert Systems with Applications,Vol. 143, p. 113082
- Lunesu and M. Marchesi (2015). Bitcoin Spread Prediction Using Social and Web
Search Media, in UMAP Workshops
- Madan, I., Saluja, S., A. Zhao A. (2015). Automated bitcoin trading via machine learning
algorithms, URL. Matta, M, I.
- McNally, S., Roche, J., Caton, S. (2018, March). “Predicting the price of bitcoin using
machine learning”, In 2018 26th Euromicro International Conference on Parallel,
Distributed and Network-based Processing (PDP) (pp. 339-343). IEEE.
غیرقطعی برای پی شبینی قیم ت.. . CNN-LSTM 176 / ............ . ارائه یک مدل ترکیبی
- Mittal, M.,Geetha, G., (2022).Predicting Bitcoin Price using Machine Learning, 2022
International Conference on Computer Communication and Informatics (ICCCI), pp. 1-
7, doi: 10.1109/ICCCI54379.2022.9740772.
- Gal, Y., Ghahramani Z., (2016).Dropout as a bayesian approximation: representing
model uncertainty in deep learning. In: international conference on machine learning, pp
1050–1059
- Greaves, A., Au, B. (2015). Using the bitcoin transaction graph to predict the price of
bitcoin, No Data.
- Patel, M.; Tanwar, S. (2020). A deep learning-based cryptocurrency price prediction
scheme for financial institutions. J. Inf. Secur., 55, 102583.
- Peng, Y., Henrique, P., Albuquerque, M.,,(2018).The best of two worlds: Forecasting
high frequency volatility for cryptocurrencies and traditional currencies with Support
Vector Regression, Expert Systems with Applications, Vol. 97, 177-192. doi:
10.1016/j.eswa.2017.12.004
- Sharifi, A. M., Khalili Damghani, K., Abdi, F., Sardar, S. (2022). 'A hybrid model for
predicting bitcoin price using machine learning and metaheuristic algorithms', Journal of
Applied Research on Industrial Engineering, 9(1), pp. 134-150. doi:
10.22105/jarie.2021.291175.1343
- Sin, E., Wang, L., (2017, July). “Bitcoin price prediction using ensembles of neural
networks”, In 2017 13th International Conference on Natural Computation, Fuzzy
Systems and Knowledge Discovery (ICNC-FSKD) (pp. 666-671). IEEE
- Srivastava, N., (2013). Improving Neural Networks with Dropout, Master, Toronto,
Canada, 2013.
- Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.
(2014).Dropout: a simple way to prevent neural networks from overfitting, J Mach Learn
Res 15(1):1929–1958
- Sovbetov, Y. (2018). Factors influencing cryptocurrency prices: Evidence from bitcoin,
ethereum, dash, litcoin, and monero, Journal of Economics and Financial Analysis, 2(2),
1-27.
- Xiong, L., Lu, Y. (2017, April). “Hybrid ARIMA-BPNN model for time series
prediction of the Chinese stock market”, In 2017 3rd International Conference on
Information Management (ICIM) (pp. 93-97). IEEE
