Development of financial and marketing models using social sentiment

First, we proposed a method for predicting the increase of Nikkei VI using postings from Yahoo! Stock BBS. The prediction model was developed using topic model and machine learning. We developed volatility trading simulation to verify the validity of the model. As a result, we confirmed a significant improvement in profit. Second, we analyzed Twitter as a political communication using social networks and natural language processing. As a result, we might find announcement effects. Also, we found that each supporter formed a group, and there were few retweets between them. In other words, Twitter is not a place to change their own opinion, but to reinforce their opinion.