Forecasting Bitcoin Price Fluctuation by Twitter Sentiment Analysis
Otabek Sattarov, Heung Seok Jeon, Ryum-Duck Oh, Jun Dong Lee
Abstract
Bitcoin is one of the main phenomena in recent times together with other cryptocurrencies due to the redefinition of the money term and its price fluctuations. Moreover, scientists are increasingly recognizing Twitter's predictive power for a wide range of events, and particularly for financial markets. This article examines to what degree Bitcoin returns can be estimated using public opinion on Twitter. Using a sentiment analyzer on Bitcoin-related tweets and financial data, the Twitter sentiment was found to have predictive power for Bitcoin's results. Once again, our findings confirm the presence of a correlation between them. We observed 62.48% accuracy when making predictions based on bitcoin-related tweet sentiment and historical bitcoin price.