Tree-bank Grammars
作者： Eugene Charniak
DOI:
关键词:
摘要: By a "tree-bank grammar" we mean context-free grammar created by reading the production rules directly from hand-parsed sentences in tree bank. Common wisdom has it that such grammars do not perform & though know of no published data on issue. The primary purpose this paper is to show common wrong. In particular, present results tree-bank based Penn WaII Street Journal To best our knowledge, outperforms ah other non-word-based statistical parsers/grammars corpus. That is, parsers consider input as string tags and ignore actual words
aaaipress.org参考文章(9)
Eugene Charniak, Sharon A. Caraballo, New figures of merit for best-first probabilistic chart parsing Computational Linguistics. ,vol. 24, pp. 275- 298 ,(1998)
Mitch Marcus, Beatrice Santorini, Mary Ann Marcinkiewicz, None, Building a large annotated corpus of English: the penn treebank Computational Linguistics. ,vol. 19, pp. 313- 330 ,(1993) , 10.21236/ADA273556
Eugene Charniak, Parsing with Context-Free Grammars and Word Statistics Brown University. ,(1995)
Eugene Charniak, Statistical Language Learning ,(1994)
Eric Brill, Automatic grammar induction and parsing free text Proceedings of the 31st annual meeting on Association for Computational Linguistics -. pp. 259- 265 ,(1993) , 10.3115/981574.981609
David M. Magerman, Statistical decision-tree models for parsing Proceedings of the 33rd annual meeting on Association for Computational Linguistics -. pp. 276- 283 ,(1995) , 10.3115/981658.981695
R. Harald Baayen, Eugene Charniak, Statistical Language Learning Language. ,vol. 73, pp. 588- ,(1997) , 10.2307/415888
Fernando Pereira, Yves Schabes, Inside-outside reestimation from partially bracketed corpora Proceedings of the 30th annual meeting on Association for Computational Linguistics -. pp. 128- 135 ,(1992) , 10.3115/981967.981984
Eugene Charniak, Sharon A. Caraballo, Figures of Merit for Best-First Probabilistic Chart Parsing empirical methods in natural language processing. ,(1996)