TY - GEN
T1 - Event extraction as dependency parsing for BioNLP 2011
AU - McClosky, David
AU - Surdeanu, Mihai
AU - Manning, Christopher D.
N1 - Funding Information:
We would like to thank the BioNLP shared task organizers for an enjoyable and interesting task and their quick responses to questions. We would also like to thank Sebastian Riedel for many interesting discussions. We gratefully acknowledge the support of the Defense Advanced Research Projects Agency (DARPA) Machine Reading Program under Air Force Research Laboratory (AFRL) prime contract no. FA8750-09-C-0181.
Publisher Copyright:
© 2011 Association for Computational Linguistics.
PY - 2011
Y1 - 2011
N2 - We describe the Stanford entry to the BioNLP 2011 shared task on biomolecular event extraction (Kim et al., 2011a). Our framework is based on the observation that event structures bear a close relation to dependency graphs. We show that if biomolecular events are cast as these pseudosyntactic structures, standard parsing tools (maximum-spanning tree parsers and parse rerankers) can be applied to perform event extraction with minimum domain-specific tuning. The vast majority of our domain-specific knowledge comes from the conversion to and from dependency graphs. Our system performed competitively, obtaining 3rd place in the Infectious Diseases track (50.6% f-score), 5th place in Epigenetics and Post-translational Modifications (31.2%), and 7th place in Genia (50.0%). Additionally, this system was part of the combined system in Riedel et al. (2011) to produce the highest scoring system in three out of the four event extraction tasks.
AB - We describe the Stanford entry to the BioNLP 2011 shared task on biomolecular event extraction (Kim et al., 2011a). Our framework is based on the observation that event structures bear a close relation to dependency graphs. We show that if biomolecular events are cast as these pseudosyntactic structures, standard parsing tools (maximum-spanning tree parsers and parse rerankers) can be applied to perform event extraction with minimum domain-specific tuning. The vast majority of our domain-specific knowledge comes from the conversion to and from dependency graphs. Our system performed competitively, obtaining 3rd place in the Infectious Diseases track (50.6% f-score), 5th place in Epigenetics and Post-translational Modifications (31.2%), and 7th place in Genia (50.0%). Additionally, this system was part of the combined system in Riedel et al. (2011) to produce the highest scoring system in three out of the four event extraction tasks.
UR - http://www.scopus.com/inward/record.url?scp=84859316201&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84859316201&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84859316201
T3 - Proceedings of BioNLP Shared Task 2011 Workshop at the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, ACL HLT 2011
SP - 41
EP - 45
BT - Proceedings of BioNLP Shared Task 2011 Workshop at the 49th Annual Meeting of the Association for Computational Linguistics
A2 - Tsujii, Jun'ichi
A2 - Kim, Jin-Dong
A2 - Pyysalo, Sampo
PB - Association for Computational Linguistics (ACL)
T2 - BioNLP Shared Task 2011 Workshop at the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, ACL HLT 2011
Y2 - 24 June 2011
ER -