Treat task as finding nodes (events, entities) and edges (args) in a graph.
Trigger detection: SVM word tagger; known multiword triggers from training are postprocessed; for cases where tokens have multiple annotations, create new double-annotation classes; postprocessing handles multiple events of the same type sharing a trigger. Large feature space.
Introduce β multiplier for negative class to calibrate precision-recall tradeoff for whole system.
Multi-class SVM for argument classification with many features, based on *groupings* of tokens along the shortest dependency path between candidate trigger and arg. Each pair-classification decision is independent.
Rule-based post-processing on graph produces valid events.
Had considered N-best reranking of candidate graphs, but couldn't build a system of any effect (it has potential of 11.5% F-score improvement).
System errors almost evenly split between trigger and edge detectors.
Ignore multi-sentence annotations, since 95% of all events are in 1 sentence.
Intend to open-source their system.
%0 Conference Paper
%1 bjorne2009
%A Björne, Jari
%A Heimonen, Juho
%A Ginter, Filip
%A Airola, Antti
%A Pahikkala, Tapio
%A Salakoski, Tapio
%B Proceedings of the BioNLP 2009 Workshop Companion Volume for Shared Task
%C Boulder, Colorado
%D 2009
%I Association for Computational Linguistics
%K biomedical bionlp09 corpus_genia event_extraction opensource
%P 10--18
%T Extracting Complex Biological Events with Rich Graph-Based Feature Sets
%U http://www.aclweb.org/anthology/W09-1402
@inproceedings{bjorne2009,
added-at = {2009-10-22T15:06:46.000+0200},
address = {Boulder, Colorado},
author = {Bj\"{o}rne, Jari and Heimonen, Juho and Ginter, Filip and Airola, Antti and Pahikkala, Tapio and Salakoski, Tapio},
biburl = {https://www.bibsonomy.org/bibtex/24b2f7bac540367271114f939e0474481/jnothman},
booktitle = {Proceedings of the BioNLP 2009 Workshop Companion Volume for Shared Task},
description = {Treat task as finding nodes (events, entities) and edges (args) in a graph.
Trigger detection: SVM word tagger; known multiword triggers from training are postprocessed; for cases where tokens have multiple annotations, create new double-annotation classes; postprocessing handles multiple events of the same type sharing a trigger. Large feature space.
Introduce β multiplier for negative class to calibrate precision-recall tradeoff for whole system.
Multi-class SVM for argument classification with many features, based on *groupings* of tokens along the shortest dependency path between candidate trigger and arg. Each pair-classification decision is independent.
Rule-based post-processing on graph produces valid events.
Had considered N-best reranking of candidate graphs, but couldn't build a system of any effect (it has potential of 11.5% F-score improvement).
System errors almost evenly split between trigger and edge detectors.
Ignore multi-sentence annotations, since 95% of all events are in 1 sentence.
Intend to open-source their system.},
interhash = {124dbb1964c59d19eaaf5e360fa65c9e},
intrahash = {4b2f7bac540367271114f939e0474481},
keywords = {biomedical bionlp09 corpus_genia event_extraction opensource},
month = {June},
pages = {10--18},
publisher = {Association for Computational Linguistics},
timestamp = {2009-10-22T15:06:46.000+0200},
title = {Extracting Complex Biological Events with Rich Graph-Based Feature Sets},
url = {http://www.aclweb.org/anthology/W09-1402},
year = 2009
}