Twitter provides in real time a huge volume of data on people's opinions towards various issues. One natural question to ask is what we can achieve with this data. Another question is how reliable are existing Twitter-based methods to gauge people's opinions.
Towards answering these two questions we have collected datasets on movies and political elections. In our first study we showed that there is indeed a discrepancy between the Twitter population and other online populations regarding movies. This suggests a need to adjust for potential bias in using Twitter data. In our second study we proposed an optimization framework to quantify political leaning of Twitter users, and implemented an automated method which is validated with data from the recent U.S. presidential election.
(music from link)
 F. M. F. Wong, S. Sen, and M. Chiang, "Why watching movie tweets won't tell the whole story?" WOSN 2012. (pdf)
 F. M. F. Wong, C. W. Tan, S. Sen, and M. Chiang, "Quantifying political leaning from tweets and retweets," ICWSM 2013. (pdf)