A Study on Machine-Learning-Based Prediction for Bitcoin’s Price via Using LSTM and SVR
Xia Zhang, Weimin Qi, Zhiming Zhan · Journal of Physics Conference Series · 2021
Abstract Bitcoin plays an increasingly crucial role currently, whose of prices are challenged to be predicted when compared to the index of any traditional stock markets. With the accent of machine learning and deep learning, it is urgent to find out whether a machine learning model can or cannot see the way that the changes of Bitcoin’s price or volume interact over some period of time leads to a price increase or decrease within next few days. Testing results indicated that, the proposed two machine learning models built by LSTM and SVR were capable to capture longer-range dependencies and performed accurately by producing state-of-the-art results.