We work on problems in web search, information retrieval and text and data mining, including:
- Web-scale text mining, information extraction, and question answering: how to organize and retrieve the information "buried'' in the millions of unstructured and semi-structured documents on the web.
- Mining user interactions and behavior information in web search: how to use the "wisdom of crowds'' -- that is, interactions of millions of searchers with the search engines -- to improve web search.
- Searching and mining social media: how to retrieve relevant and reliable information from social media, such as community question answering sites.
- Medical informatics: in particular, managing unstructured in formation in Electronic Medical Records
