DMCA
Two Strong Baselines for the BioNLP 2009 Event Extraction Task
| Venue: | Proceedings of the 2010 Workshop on Biomedical Natural Language Processing |
| Citations: | 3 - 0 self |
BibTeX
@INPROCEEDINGS{Vlachos_twostrong,
author = {Andreas Vlachos},
title = {Two Strong Baselines for the BioNLP 2009 Event Extraction Task},
booktitle = {Proceedings of the 2010 Workshop on Biomedical Natural Language Processing},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
This paper presents two strong baselines for the BioNLP 2009 shared task on event extraction. First we re-implement a rulebased approach which allows us to explore the task and the effect of domainadapted parsing on it. We then replace the rule-based component with support vector machine classifiers and achieve performance near the state-of-the-art without using any external resources. The good performances achieved and the relative simplicity of both approaches make them reproducible baselines. We conclude with suggestions for future work with respect to the task representation. 1
