Twitter Sentiment Analysis of Bitcoin Price Fluctuation with Machine Learning Techniques

The term ‘currency’ and its volatile day to day price movements have not only brought bitcoin out in the open but also the other cryptocurrencies. Scientists, though, are also aware that Twitter can be predictive, particularly in the financial markets. In this research article, we test the ability of the public Twitter opinion to influence Bitcoin returns. It was found using Textblob on Bitcoin related tweets that Twitter sentiment is capable of predicting Bitcoin results. Also, get to compare results using different machine learning such as Naïve Bayes, Support Vector machine, Logistic Regression and Random Forest to arrive to accurate prediction on bitcoin related tweets. It was found that the results of the Naïve Bayes are better than that of other algorithms.