Paper: Two Strong Baselines for the BioNLP 2009 Event Extraction Task

ACL ID W10-1901
Title Two Strong Baselines for the BioNLP 2009 Event Extraction Task
Venue Workshop on Biomedical Natural Language Processing
Session
Year 2010
Authors

This paper presents two strong baselines for the BioNLP 2009 shared task on event extraction. First we re-implement a rule- based approach which allows us to ex- plore the task and the effect of domain- adapted parsing on it. We then replace the rule-based component with support vec- tor machine classifiers and achieve perfor- mance near the state-of-the-art without us- ing any external resources. The good per- formances achieved and the relative sim- plicity of both approaches make them re- producible baselines. We conclude with suggestions for future work with respect to the task representation.