Summary

From scientific journals to newspapers, weblogs and email, text is a rich and expanding source of information. Language Technology (LT) aims to provide tools for mining this source and delivering text-based information to the people who need it -- scientists, business and industry analysts, intelligence agents, etc.

Over time, Language Technology has been moving up the food chain, developing and exploiting techniques based on words, then word sequences (N-grams), syntactic structures, and most recently, dependency relations. Still to come is the higher-order information available from the sequences of sentences that make up discourse.

To show what is needed to recognize and extract such information, I will start by reviewing the features of discourse that trigger the additional, higher-order meaning it conveys. A theoretical account of these triggers is provided in a lexicalised grammar for discourse called DLTAG. Detailed characteristics of these triggers, enabling their use in LT will be provided by the "Penn Discourse TreeBank", which saw its first release in March 2006 (http://www.seas.upenn.edu/~pdtb/), with a full release planned for April 2007. I will conclude the talk by describing features of this resource, what it might allow researchers to discover, and how such discoveries could enhance Language Technology.

Bio

Bonnie Webber received her PhD from Harvard University and taught at the University of Pennsylvania (Department of Computer & Information Science) for 20 years. She is now a professor in the School of Informatics at the University of Edinburgh.

She has carried out and supervised research on Question Answering (starting with BBN's LUNAR system in the early 70's), discourse phenomena (starting with her PhD thesis on discourse anaphora), animation from instructions, medical decision support systems and (more recently) bioinformatics. She has recently published papers in the journals Computational Linguistics, Cognitive Science, and the Journal of Semantics.