Paper: A Best-First Probabilistic Shift-Reduce Parser

ACL ID P06-2089
Title A Best-First Probabilistic Shift-Reduce Parser
Venue Annual Meeting of the Association of Computational Linguistics
Session Poster Session
Year 2006

Recently proposed deterministic classifier- based parsers (Nivre and Scholz, 2004; Sagae and Lavie, 2005; Yamada and Mat- sumoto, 2003) offer attractive alternatives to generative statistical parsers. Determin- istic parsers are fast, efficient, and sim- ple to implement, but generally less ac- curate than optimal (or nearly optimal) statistical parsers. We present a statis- tical shift-reduce parser that bridges the gap between deterministic and probabilis- tic parsers. The parsing model is essen- tially the same as one previously used for deterministic parsing, but the parser performs a best-first search instead of a greedy search. Using the standard sec- tions of the WSJ corpus of the Penn Tree- bank for training and testing, our parser has 88.1% precision and 87.8% recall (us- ing auto...