TweetSmart: Hedging in markets through Twitter

摘要: Application of pattern recognition and machine learning in highly dynamic data intensive financial markets is a popular research area amongst researchers analysts. With evolving social dynamics millions across the globe, it provides opportunity to make use patterns investor sentiment comprising large scale microblog discussions understand market movements an effective application making hedging decisions. We apply analysis principles study causation between public collective movements. In this work we have used 0.6 million tweets for period November 2010 June 2011, run practical simulation model Dow Jones Industrial Average-DJIA Index. elaborated on how simple strategy like married-put can exercise weekly directional forecasts DJIA portfolio adjustments from risky high conditions vice versa. found maximum 91% SVM based binary classifier accuracy, towards direction (up down prediction) estimations DJIA.