Browse > Article
http://dx.doi.org/10.15207/JKCS.2022.13.04.017

A Study on the Hyper-parameter Optimization of Bitcoin Price Prediction LSTM Model  

Kim, Jun-Ho (Department of Game Design and Develpment, Sangmyung University)
Sung, Hanul (Department of Game Design and Develpment, Sangmyung University)
Publication Information
Journal of the Korea Convergence Society / v.13, no.4, 2022 , pp. 17-24 More about this Journal
Abstract
Bitcoin is a peer-to-peer cryptocurrency designed for electronic transactions that do not depend on the government or financial institutions. Since Bitcoin was first issued, a huge blockchain financial market has been created, and as a result, research to predict Bitcoin price data using machine learning has been increasing. However, the inefficient Hyper-parameter optimization process of machine learning research is interrupting the progress of the research. In this paper, we analyzes and presents the direction of Hyper-parameter optimization through experiments that compose the entire combination of the Timesteps, the number of LSTM units, and the Dropout ratio among the most representative Hyper-parameter and measure the predictive performance for each combination based on Bitcoin price prediction model using LSTM layer.
Keywords
Bitcoin; Cryptocurrency; LSTM; Deep-learning; Data Prediction; Optimization;
Citations & Related Records
Times Cited By KSCI : 3  (Citation Analysis)
연도 인용수 순위
1 S. McNally, J. Roche & S. Caton. (2018). Predicting the Price of Bitcoin Using Machine Learning. 26th Euromicro International Conference on Parallel, Distributed, and Network-Based Processing, 339-343.
2 T. Yu & H. Zhu. (2020). Hyper-Parameter Optimization: A Review of Algorithms and Applications.
3 E. M. Kim & J. W. Hong. (2021). The Prediction model of cryptocurrency price using news sentiment analysis and deep learning. Knowledge Management Review, 22(2), 19-32.   DOI
4 M. G. Gang, B. S. Kim, M. G. Shin, U. J. Baek & M. S. Kim. (2020). LSTM-based Prediction of Bitcoin Price Fluctuation using Sentiment Analysis. The Fall Conference of The Korean Institute of Communications and Information Sciences Conference, 1-2.
5 G. Sui, Y. Yu. (2020). Bayesian Contextual Bandits for Hyper Parameter Optimization. IEEE Access, 8, 42971-42979.   DOI
6 D. H. Kown, J. S. Heo, J. B. Kim, H. K. Lim & Y. H. Han (2018). Correlation Analysis and Regression Test on Cryptocurrency Price Data. Proceedings of the Korea Information Processing Society Conference, 346-349.
7 https://www.binance.com
8 I. T. Joo & S. H. Choi. (2018). Stock Prediction Model based on Bidirectional LSTM Recurrent Neural Network. Journal of Korea Institute of Information, Electronics, and Communication Technology 11(2), 204-208.   DOI
9 S. W. Kim. (2021). Performance Analysis of Bitcoin Investment Strategy using Deep Learning. Journal of the Korea Convergence Society, 12(4), 249-258.   DOI
10 J. G. Won & T. H. Hong. (2021). The Prediction of Cryptocurrency on Using Text Mining and Deep Learning Techniques : Comparison of Korean and USA Market. Knowledge Management Research, 22(2), 1-17.
11 K. J. Jeong & J. S. Choi. (2015). Deep Recurrent Neural Network. Communications of the Korean Institute of Information Scientists and Engineers, 33(8), 39-43.
12 https://upbit.com
13 M. G. Gang, B. S. Kim & H. J. Shin. (2021). Machine Learning based Prediction of Bitcoin Price Fluctuation Using News Data. Korean Network Operations and Management Conference, 38-39.
14 S. H. Ji, U. J. Baek, M. G. Shin, J. S. Park & M. S. Kim. (2019). A Study on the Prediction of Number of Bitcoin Network Transactions Based on Machine Learning. Korean Network Operations and Management Review, 22(1), 68-76.
15 J. Bergstra & Y. Bengio. (2012). Random search for hyper-parameter optimization. Journal of Machine Learning Resarch, 13(2).
16 Y. U. Choi, D. U. Yoon, J. H. Choi & J. M. Byun. (2020). Hyperparameter Search for Facies Classification with Bayesian Optimization. Geophysics and Geophysical Exploration, 23(3), 157-167.   DOI
17 S. H. Ji, U. J. Baek, M. G. Shin, Y. H. Goo, S. H. Yoon & M. S. Kim. (2019). Design of LSTM Model for Prediction of Number of Bitcoin Transactions. Proceedings of Symposium of the Korean Institute of communications and Information Sciences, 472-473.