Our group is involved in research at the intersection of machine learning and computational linguistics. Our philosophy is centered around the importance of clear semantic and functional representations of the knowledge to be extracted from linguistic structures and rigorous use of annotated corpora to that end.
We are particularly interested in examining how machine learning techniques can help to capture large-scale linguistic properties of texts that capture levels of meaning above those available at the word or clause level, paying specific attention to elucidating the relationships among linguistic structures, individual reasoning, and social context.
A focus of ours is work on extracting and using non-referential semantic properties of text, such as rhetorical organization or style.