Transformational Priors Over Grammars (2002)
Cached
Download Links
| Venue: | PROCEEDINGS OF THE CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING (EMNLP 2002), PHILADELPHIA |
| Citations: | 6 - 1 self |
BibTeX
@INPROCEEDINGS{Eisner02transformationalpriors,
author = {Jason Eisner},
title = {Transformational Priors Over Grammars},
booktitle = {PROCEEDINGS OF THE CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING (EMNLP 2002), PHILADELPHIA},
year = {2002},
pages = {63--70},
publisher = {}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
This paper proposes a novel class of PCFG parameterizations that support linguistically reasonable priors over PCFGs. To estimate the parameters is to discover a notion of relatedness among context-free rules such that related rules tend to have related probabilities. The prior favors grammars in which the relationships are simple to describe and have few major exceptions. A basic version that bases relatedness on weighted edit distance yields superior smoothing of grammars learned from the Penn Treebank (20% reduction of rule perplexity over the best previous method).
