Sentiment analysis on tweets in a financial domain

This paper investigates whether sentiment analysis of public mood derived from large-scale Twitter feeds can be used to identify important events and predict movements of stock prices. We used the volume and sentiment polarity of Apple financial tweets to identify important events and predict future movements of Apple stock prices. Statistical analysis using the Granger causality test showed that we were able to predict the rise or fall in closing price of Apple stocks two days before the change happens.