Transition network grammars for natural language analysis
Communications of the ACM
Head-driven statistical models for natural language parsing
Head-driven statistical models for natural language parsing
Robustness beyond shallowness: incremental deep parsing
Natural Language Engineering
Constraint based integration of deep and shallow parsing techniques
EACL '03 Proceedings of the tenth conference on European chapter of the Association for Computational Linguistics - Volume 1
Integrated shallow and deep parsing: TopP meets HPSG
ACL '03 Proceedings of the 41st Annual Meeting on Association for Computational Linguistics - Volume 1
ACL '04 Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics
Reranking and self-training for parser adaptation
ACL-44 Proceedings of the 21st International Conference on Computational Linguistics and the 44th annual meeting of the Association for Computational Linguistics
Using an incremental robust parser to automatically generate semantic UNL graphs
ROMAND '04 Proceedings of the 3rd Workshop on RObust Methods in Analysis of Natural Language Data
TextTree construction for parser and treebank development
Software '05 Proceedings of the Workshop on Software
A classifier-based parser with linear run-time complexity
Parsing '05 Proceedings of the Ninth International Workshop on Parsing Technology
Symbolic preference using simple scoring
IWPT '07 Proceedings of the 10th International Conference on Parsing Technologies
Grammaticality judgement in a word completion task
CL&W '10 Proceedings of the NAACL HLT 2010 Workshop on Computational Linguistics and Writing: Writing Processes and Authoring Aids
Hi-index | 0.00 |
Contemporary parser research is, to a large extent, focused on statistical parsers and deep-unification-based parsers. This paper describes an alternative, hybrid architecture in which an ATN-like parser, augmented by many preference tests, builds on the results of a fast chunker. The combination is as efficient as most stochastic parsers, and accuracy is close and continues to improve. These results raise questions about the practicality of deep unification for symbolic parsing.