Procedure for quantitatively comparing the syntactic coverage of English grammars
HLT '91 Proceedings of the workshop on Speech and Natural Language
Statistical parsing of messages
HLT '90 Proceedings of the workshop on Speech and Natural Language
The acquisition and use of context-dependent grammars for English
Computational Linguistics
New figures of merit for best-first probabilistic chart parsing
Computational Linguistics
Automatic compensation for parser figure-of-merit flaws
ACL '99 Proceedings of the 37th annual meeting of the Association for Computational Linguistics on Computational Linguistics
Hi-index | 0.00 |
We describe a series of three experiments in which supervised learning techniques were used to acquire three different types of grammars for English news stories. The acquired grammar types were: 1) context-free, 2) context-dependent, and 3) probabilistic context-free. Training data were derived from University of Pennsylvania Treebank parses of 50 Wall Street Journal articles. In each case, the system started with essentially no grammatical knowledge, and learned a set of grammar rules exclusively from the training data. Performance for each grammar type was then evaluated on an independent set of test sentences using Parseval, a standard measure of parsing accuracy. These experimental results yield a direct quantitative comparison between each of the three methods.