Forecasting Stock Prices using Social Media Analysis

From REU@MU
Jump to: navigation, search

Last year, Scott Coyne developed a prediction model that used twits from stocktwits.com to predict the rise and fall of the stock market. This model used textual analysis, linear regression and smart user identification to generate predictions. My goal is to investigate a way to identify bots in the social media system, implement different machine learning algorithms, and/or improve upon the textual analysis system.