Analysis On Relationship Between Bitcoin Price Trend And Sentiment Of Bitcoin Related Tweets By ML And NLP

Abstract

The analysis has been shown to be useful in predicting whether the price of Bitcoin will rise or fall. However, current technology can only predict price direction, not magnitude of increase/decrease. In this project, we seek to build on the state-of-the-art to not only predict the direction yet to also predict the magnitude of increase/decrease. We present results from experiments that investigated the relationship between sentiment and future price at various temporal granularities, with the goal of determining the optimal time interval at which the expressed sentiment becomes a reliable predictor of price change. This study examines the relationship between bitcoin-related tweets and bitcoin price from January 2019 to May 2019, using Natural Language Processing (NLP) and Machine Learning (ML).