Paper: Analysis of Discourse Structure with Syntactic Dependencies and Data-Driven Shift-Reduce Parsing

ACL ID W09-3813
Title Analysis of Discourse Structure with Syntactic Dependencies and Data-Driven Shift-Reduce Parsing
Venue International Conference on Parsing Technologies
Session Main Conference
Year 2009
Authors
  • Kenji Sagae (University of Southern California, Marina del Rey CA)

We present an eficient aproach for dis- course parsing within and acros sen- tences, where the unit of procesing is an entire document, and not a single sen- tence. We aply shift-reduce algorithms for dependency and constituent parsing to determine syntactic dependencies for the sentences in a document, and subse- quently a Rhetorical Structure Theory (RST) tre for the entire document. Our results show that our linear-time shift- reduce framework achieves high acu- racy and a large improvement in efi- ciency compared to a state-of-the-art ap- proach based on chart parsing with dy- namic programing.