Learning Probabilistic Dependency Grammars from Labeled Text

Glenn Carroll and Eugene Charniak

We present the results of experimenting with schemes for learning probabilistic dependency grammars for English from corpora labelled with part-of-speech information. We intend our system to produce widecoverage grammars which have some resemblance to the standard 2 context-free grammars of English which grammarians and linguists commonly exhibit as exampies.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.