Main projects


Opinion Extraction : Can computers learn to recognize and interpret opinions, beliefs, and sentiment from text?

I developed techniques to improve the automatic extraction of such information via joint inference of opinion expressions, opinion holders, opinion targets as well as their association relations from text.




Figure 1: Examples of the system extracted opinions. The highlighted text spans indicate the extracted opinion expressions, opinion holders and opinion targets, and the labeled arrows indicate their association relations. The green color indicates positive polarity, red indicates negative polarity, orange indicates neutrality, and ★ indicates intensity (more stars means higher intensity).


Publications


Event Coreference : Can computers learn to detect identical events in a large corpus of news articles?

I developed a hierarchical distance-dependent Bayesian clustering model for within- and cross-document event coreference resolution.

Figure 2: Examples of the system extracted coreferential events. The highlighted text spans indicate the extracted event mentions (triggers) and event arguments (participants, times and locations). Two event mentions are coreferent if they (along with their contexts) refer to the same actual event.


Publications