This paper aims to prove whether Twitter data relating to cryptocurrencies can be utilized to develop advantageous crypto coin trading strategies. By way of supervised machine learning techniques, our team will outline several machine learning pipelines with the objective of identifying cryptocurrency market movement. Our approach to cleaning data and applying supervised learning algorithms such as logistic regression, Naive Bayes, and support vector machines leads to a final hour-to-hour prediction accuracy exceeding 90%. In order to achieve this result, rigorous error analysis is employed in order to ensure that accurate inputs are utilized at each step.
Auto74
http://cs229.stanford.edu/proj2015/029_report.pdf