The Technology

TweetTrader.net is maintained by a team at the TUM School of Management and leverages the latest results from research at the crossroads of finance and natural language processing.

The focus is on the exploration of Twitter as an indicator of financial market activity. We are using dictionary-based approaches and machine learning techniques to classify message board content automatically and extract the sentiment contained in the postings. Research results shows that our technology to measure market sentiment can detect statistically significant abnormal returns.