Prediction of Bitcoin Price through LSTM, ARIMA, XGBoost, Prophet and Sentiment Analysis on Dynamic Streaming Data
K. Ramani, M. Jahnavi, P. Jagadeesh Reddy, P. VenkataChakravarthi, P. Meghanath, S. K. Imran
Abstract
The most popular cryptocurrency in the world is Bitcoin, which enables users to perform secure online transactions. When carrying out quick transactions, including cash transactions, this aids in keeping your money secret. Most of Consumers have been interested in the Bitcoin ecosystem in recent years. Predicting the bitcoin price accurately is a difficult task due to its high volatility. In this paper, we used deep learning and machine learning algorithms namely Long Short-Term Memory, Autoregressive Integrated Moving Average , XGBoost, Prophet and Sentiment analysis were performed on bitcoin data.The algorithms were trained on live streaming finanacial data, and results are compared based on predicted metrics like Root mean Square Error,Mean Absolute Error and R2. The results show that Sentiment analysis combined with LSTM provide better performance in bitcoin price prediction of all other algorithms.