Flexible margin selection for reranking with full pairwise samples (2004)
| Venue: | In Proc. of 1st IJCNLP |
| Citations: | 11 - 2 self |
BibTeX
@INPROCEEDINGS{Shen04flexiblemargin,
author = {Libin Shen and Aravind K. Joshi},
title = {Flexible margin selection for reranking with full pairwise samples},
booktitle = {In Proc. of 1st IJCNLP},
year = {2004},
pages = {446--455}
}
Years of Citing Articles
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
Perceptron like large margin algorithms are introduced for the experiments with various margin selections. Compared to the previous perceptron reranking algorithms, the new algorithms use full pairwise samples and allow us to search for margins in a larger space. Our experimental results on the data set of (Collins, 2000) show that a perceptron like ordinal regression algorithm with uneven margin can achieve Recall/Precision of 89.5/90.0 on section 23 of WSJ PTB. Our result on margin selection can be employed in other large margin machine learning algorithms as well as in other NLP tasks. 1
Citations
No citations identified.
