A Markov Logic Approach to Bio-Molecular Event Extraction
S. Riedel, H. Chun, T. Takagi, and J. Tsujii. Proceedings of the BioNLP 2009 Workshop Companion Volume for Shared Task, page 41--49. Boulder, Colorado, Association for Computational Linguistics, (June 2009)
Description
Learn a joint probabilistic model (not pipeline) with Markov Logic (Statistical Relation Learning language). 3-4th in BioNLP; best results for task 2.
Represent events as relational structures over the tokens of sentence (close to SRL). Essentially First-Order Logic formulae with weights. Markov Logic does not yet handle cases where the number and identity of entities is unknown.
Map event annotations into MLNs: E = (L, C) where L is a set of token-token links, and C is a set of event triggers.
Four hidden predicates (like target class features): event(i), eventType(i, t), site(i), role(i, j, r)
Other predicates (features) regard entity annotations, dependency paths (from Charniak and CCG), dictionary lookups.
Provide implication logic for features to target hidden predicates (each gets assigned a weight in training), and global constraints. Many global constraints help validate; additional ones are in response to error analysis (e.g. token cannot be argument of multiple events).
Note that constraints on role assignment can improve event trigger detection in a joint model.
%0 Conference Paper
%1 riedel2009
%A Riedel, Sebastian
%A Chun, Hong-Woo
%A Takagi, Toshihisa
%A Tsujii, Jun'ichi
%B Proceedings of the BioNLP 2009 Workshop Companion Volume for Shared Task
%C Boulder, Colorado
%D 2009
%I Association for Computational Linguistics
%K bionlp09 dependency_parse event_extraction joint_model markov_logic
%P 41--49
%T A Markov Logic Approach to Bio-Molecular Event Extraction
%U http://www.aclweb.org/anthology/W09-1406
@inproceedings{riedel2009,
added-at = {2009-10-22T15:17:55.000+0200},
address = {Boulder, Colorado},
author = {Riedel, Sebastian and Chun, Hong-Woo and Takagi, Toshihisa and Tsujii, Jun'ichi},
biburl = {https://www.bibsonomy.org/bibtex/25c440785528f573f3ea831c2a0d29606/jnothman},
booktitle = {Proceedings of the BioNLP 2009 Workshop Companion Volume for Shared Task},
description = {Learn a joint probabilistic model (not pipeline) with Markov Logic (Statistical Relation Learning language). 3-4th in BioNLP; best results for task 2.
Represent events as relational structures over the tokens of sentence (close to SRL). Essentially First-Order Logic formulae with weights. Markov Logic does not yet handle cases where the number and identity of entities is unknown.
Map event annotations into MLNs: E = (L, C) where L is a set of token-token links, and C is a set of event triggers.
Four hidden predicates (like target class features): event(i), eventType(i, t), site(i), role(i, j, r)
Other predicates (features) regard entity annotations, dependency paths (from Charniak and CCG), dictionary lookups.
Provide implication logic for features to target hidden predicates (each gets assigned a weight in training), and global constraints. Many global constraints help validate; additional ones are in response to error analysis (e.g. token cannot be argument of multiple events).
Note that constraints on role assignment can improve event trigger detection in a joint model.},
interhash = {9c37cf0a5474376a28a633f41945dedd},
intrahash = {5c440785528f573f3ea831c2a0d29606},
keywords = {bionlp09 dependency_parse event_extraction joint_model markov_logic},
month = {June},
pages = {41--49},
publisher = {Association for Computational Linguistics},
timestamp = {2009-10-22T15:17:55.000+0200},
title = {A Markov Logic Approach to Bio-Molecular Event Extraction},
url = {http://www.aclweb.org/anthology/W09-1406},
year = 2009
}