Strictly lexical dependency parsing
- Qin Iris Wang, Department of Computing Science, University of Alberta
- Dale Schuurmans, AICML
- Dekang Lin
We present a strictly lexical parsing model where all the parameters are based on the words. This model does not rely on part-of-speech tags or grammatical categories. It maximizes the conditional probability of the parse tree given the sentence. This is in contrast with most previous models that compute the joint probability of the parse tree and the sen-tence. Although the maximization of joint and conditional probabilities are theoretically equivalent, the conditional model allows us to use distributional word similarity to generalize the ob-served frequency counts in the training corpus. Our experiments with the Chi-nese Treebank show that the accuracy of the conditional model is 13.6% higher than the joint model and that the strictly lexicalized conditional model outper forms the corresponding unlexicalized model based on part of-speech tags.
Citation
Q. Wang, D. Schuurmans, D. Lin. "Strictly lexical dependency parsing". International Workshop on Parsing Technologies (IWPT), January 2005.| Keywords: | machine learning |
| Category: |
BibTeX
@incollection{Wang+al:IWPT05,
author = {Qin Iris Wang and Dale Schuurmans and Dekang Lin},
title = {Strictly lexical dependency parsing},
booktitle = {International Workshop on Parsing Technologies (IWPT)},
year = 2005,
}Last Updated: March 07, 2007Submitted by Nelson Loyola
