Shift-Reduce Constituent Parsing with Neural Lookahead Features
Citations Over TimeTop 10% of 2017 papers
Abstract
Transition-based models can be fast and accurate for constituent parsing. Compared with chart-based models, they leverage richer features by extracting history information from a parser stack, which consists of a sequence of non-local constituents. On the other hand, during incremental parsing, constituent information on the right hand side of the current word is not utilized, which is a relative weakness of shift-reduce parsing. To address this limitation, we leverage a fast neural model to extract lookahead features. In particular, we build a bidirectional LSTM model, which leverages full sentence information to predict the hierarchy of constituents that each word starts and ends. The results are then passed to a strong transition-based constituent parser as lookahead features. The resulting parser gives 1.3% absolute improvement in WSJ and 2.3% in CTB compared to the baseline, giving the highest reported accuracies for fully-supervised parsing.
Related Papers
- Parsing Schemata - a framework for specification and analysis of parsing algorithms(1997)
- → Building parsers based on syntax diagrams with multiport components(2022)1 cited
- → Multiple-Human Parsing in the Wild(2017)61 cited
- → Partial Parsing: Combining Choice with Commitment(2008)6 cited
- → Analyzing Holistic Parsers: Implications for Robust Parsing and Systematicity(2001)2 cited