Stochastic attribute-value grammars

I define stochastic attribute-value grammars & give an algorithm for computing the maximum-likelihood estimate of their parameters. The estimation algorithm is adapted from Stephen Della Pietra, Vincent Della Pietra, & John Lafferty (1995). To estimate model parameters, it is necessary to co...

Full description

Saved in:
Bibliographic Details
Published in:Computational linguistics - Association for Computational Linguistics 1997-12, Vol.23 (4), p.597-618
Main Author: ABNEY, S. P
Format: Article
Language:English
Subjects:
Applied linguistics
Computational linguistics
Linguistics
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:I define stochastic attribute-value grammars & give an algorithm for computing the maximum-likelihood estimate of their parameters. The estimation algorithm is adapted from Stephen Della Pietra, Vincent Della Pietra, & John Lafferty (1995). To estimate model parameters, it is necessary to compute the expectations of certain functions under random fields. In the application discussed by Della Pietra, Della Pietra, & Lafferty (representing English orthographic constraints), Gibbs sampling can be used to estimate the needed expectations. The fact that attribute-value grammars generate constrained languages makes Gibbs sampling inapplicable, but I show that sampling can be done using the more general Metropolis-Hastings algorithm. 4 Tables, 13 Figures, 2 Appendixes, 8 References. Adapted from the source document
ISSN:0891-2017
1530-9312