摘要：Bitcoin is a kind of cryptocurrency that has become a popular stock market investment and it has been steadily rising in recent years, and occasionally falling without warning, on the stock market. Because of its fluctuations, an automated tool for predicting bitcoin on the stock market is required. However, because of its volatility, investors will need a prediction tool to help them make investment decisions in bitcoin or other cryptocurrencies. In this paper, Deep learning mechanisms like Recurrent Neural Network (RNN) and Long short-term memory (LSTM) are proposed to develop a model to forecast the bitcoin price trend in the market. Finally, the predictions result for the Bitcoin price trend are presented over the next 15, 30, and 60 days. Each model is evaluated in terms of Mean Absolute Percentage Error (MAPE) and Root Mean Squared Error (RMSE) forecasting error values. The LSTM model is found to be the better mechanism for time-series cryptocurrency price prediction, but it takes longer to compile.
关键词：Bitcoin;Blockchain;Cryptocurrency;Long Short Term Memory(LSTM);Machine Learning;Prediction;Recurrent Neural Network (RNN)
