本論文運用長短期記憶模型(Long Short-Term Memory, LSTM) 來預測比特幣(Bitcoin)價格走向。特徵值資料包含內部及外部特徵值，各抽取自比特幣區塊鏈以及交易中心。
加密貨幣是一種新型態的貨幣，其交易運行在網路中。在所有加密貨幣中，比特幣(Bitcoin, BTC)是第一個加密貨幣，且目前擁有最高的市值。預測比特幣價格是一個新興的研究題目，因為其與傳統金融資產有所差異，且其價格非常波動。
本論文對比特幣區塊鏈資料處理方法提出指引，並將長短期記憶模型實務應用到比特幣價格預測。

This thesis focuses on applying Long Short-Term Memory (LSTM) technique to predict Bitcoin price direction. Features including internal and external features are extracted from Bitcoin blockchain and exchange center respectively.
Cryptocurrency is a new type of currency that is traded over the infrastructure of Internet. Bitcoin (BTC) is the first cryptocurrency and ranks first in the market capitalization among all the other cryptocurrencies. Predicting Bitcoin price is a novel topic because of its differences with traditional financial assets and its volatility.
As contributions, this thesis provides a guide of processing Bitcoin blockchain data and serves as an empirical study on applying LSTM to Bitcoin price prediction.

[1] S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Computation, vol. 9, no. 8, 1997.
[2] S. Nakamoto, “Bitcoin: A peer-to-peer electronic cash system,” 2009.
[3] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. MIT Press, 2016.
[4] Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, 2015.
[5] F. A. Gers, J. A. Schmidhuber, and F. A. Cummins, “Learning to forget: Continual prediction with lstm,” Neural Comput., vol. 12, no. 10, 2000.
[6] S. J. Taylor, An Introduction to Volatility. Princeton University Press, 2005.
[7] Investopedia. Volatility. [Online]. Available: https://www.investopedia.com/terms/v/volatility.asp
[8] W. Huang, Y. Nakamori, and S.-Y. Wang, “Forecasting stock market movement direction with support vector machine,” Computers & Operations Research, vol. 32, no. 10, 2005.
[9] S. A. Hamid and Z. Iqbal, “Using neural networks for forecasting volatility of sp 500 index futures prices,” Journal of Business Research, 2004.
[10] A. Vejendla and D. Enke, “Evaluation of garch, rnn and fnn models for forecasting volatility in the financial markets,” IUP Journal of Financial Risk Management, vol. 10, no. 1, 2013.
[11] R. Akita, A. Yoshihara, T. Matsubara, and K. Uehara, “Deep learning for stock prediction using numerical and textual information,” in 2016 IEEE/ ACIS 15th International Conference on Computer and Information Science (ICIS), 2016.
[12] M. Matta, M. I. Lunesu, and M. Marchesi, “Bitcoin spread prediction using social and web search media,” in UMAP Workshops, 2015.
[13] I. Madan and S. Saluja, “Automated bitcoin trading via machine learning algorithms,” Stanford University, 2014.
[14] A. Greaves and B. Au, “Using the bitcoin transaction graph to predict the price of bitcoin,” Stanford University, 2015.
[15] S. McNally, “Predicting the price of bitcoin using machine learning,” Master’s thesis, Dublin, National College of Ireland, 2016.
[16] H. Jang and J. Lee, “An empirical study on modeling and prediction of bitcoin prices with bayesian neural networks based on blockchain information,” IEEE Access, vol. 6, 2018.
[17] Y. Bengio, “Learning deep architectures for ai,” Foundations and Trends® in Machine Learning, vol. 2, no. 1, 2009.
[18] Y. Bengio, A. Courville, and P. Vincent, “Representation learning: A review and new perspectives,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 35, no. 8, 2013.
[19] J. L. Elman, “Finding structure in time,” Cognitive Science, vol. 14, no. 2, 1990.
[20] Z. C. Lipton, “A critical review of recurrent neural networks for sequence learning,” CoRR, vol. abs/1506.00019, 2015.
[21] A. Graves, Supervised Sequence Labelling with Recurrent Neural Networks. Springer-Verlag Berlin Heidelberg, 2012.
[22] K. Greff, R. K. Srivastava, J. Koutník, B. R. Steunebrink, and J. Schmidhuber, “LSTM: A search space odyssey,” CoRR, vol. abs/1503.04069, 2015.
[23] Wikipedia contributors, “Loss functions for classification — Wikipedia, the free encyclopedia,” 2018. [Online]. Available: https://en.wikipedia.org/w/index.php?title=Loss_functions_for_classification&oldid=838253245
[24] Wikipedia contributors, “Gradient descent — Wikipedia, the free encyclopedia,” 2018. [Online]. Available: https://en.wikipedia.org/w/index.php?title=Gradient_descent&oldid=845809247
[25] R. Rojas, Neural Networks: A Systematic Introduction. Berlin, Heidelberg: Springer-Verlag, 1996.
[26] J. Bergstra and Y. Bengio, “Random search for hyper-parameter optimization,” J. Mach. Learn. Res., vol. 13, 2012.
[27] N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, “Dropout: A simple way to prevent neural networks from overfitting,” Journal of Machine Learning Research, vol. 15, 2014.
[28] S. Ruder, “An overview of gradient descent optimization algorithms,” CoRR, vol. abs/1609.04747, 2016.
[29] D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” CoRR, vol. abs/1412.6980, 2014.
[30] S. Dziembowski, “Introduction to cryptocurrencies,” 2015.
[31] I. Bentov, A. Gabizon, and A. Mizrahi, “Cryptocurrencies without proof of work,” CoRR, vol. abs/1406.5694, 2014.
[32] Proof of work. [Online]. Available: https://en.bitcoin.it/wiki/Proof_of_work
[33] A. Narayanan, J. Bonneau, E. W. Felten, A. Miller, S. Goldfeder, and J. Clark, Bitcoin and Cryptocurrency Technologies. Princeton University Press, 2016.
[34] Gdax exchange center documentation. [Online]. Available: https://docs.gdax.com/
[35] blockchain.info. [Online]. Available: https://blockchain.info/
[36] T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning. Springer New York Inc., 2001.
[37] Keras. [Online]. Available: https://keras.io/
[38] Nvidia. [Online]. Available: http://www.nvidia.com/page/home.html
[39] A. Karpathy, “The unreasonable effectiveness of recurrent neural networks,” 2015. [Online]. Available: http://karpathy.github.io/2015/05/21/rnn-effectiveness/