Description

Members

Publications

Demos

Funding

Research Home

PDP
A Maximum Entropy-Based Approach to Discourse Parsing

Description

Natural language engineers interested in building robust information extraction, summarization, translation, and dialogue management systems have often acknowledged that to achieve high levels of performance they need to understand the underlying structure of texts/dialogues. Although recent developments in the field are encouraging and discourse research has already yielded a number of successful applications, current discourse parsers are still far from human performance levels. Two factors contribute to this situation.

  • First, the relationship between discourse structure and lexicogrammar is still insuficiently understood. We still do not know what lexicogrammatical features correlate with certain discourse structures and relations.
  • Second, the discourse parsing algorithms implemented to date do not take advantage of formalism and algorithms specific to information and probability theories, which have been shown to produce impressive results in the field of syntactic parsing.

The goal of the PDP project is to advance the state of the art in discourse processing by addressing these two shortcomings.

Copyright © 2003 The University of Southern California. All RIGHTS RESERVED.
All trademarks mentioned herein belong to their respective owners.
Designed and maintained by Erika Barragan-Nunez.
Generated by Bryant Huang.