Biomedical event detection using rules, conditional random fields and parse tree distances

Farzaneh Sarafraz; James Eales; R Mohammadi; Jonathan Dickerson; David Robertson; Goran Nenadic

doi:10.3115/1572340.1572359

Biomedical event detection using rules, conditional random fields and parse tree distances

Farzaneh Sarafraz, James Eales, R Mohammadi, Jonathan Dickerson, David Robertson, Goran Nenadic

Division of Cardiovascular Sciences (L5)

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

141 Downloads (Pure)

Abstract

This paper reports on a system developed for the BioNLP'09 shared task on detection and characterisation of biomedical events. Event triggers and types were recognised using a conditional random field classifier and a set of rules, while event participants were identified using a rule-based system that relied on relative distances between candidate entities and the trigger in the associated parse tree. The results on previously unseen test data were encouraging: for non-regulatory events, the F-score was almost 50% (with precision above 60%), with the overall F-score of around 30% (49% precision). The performance on more complex regulatory events was poor (F-measure of 7%). Among the 24 teams submitting the test results, our results were ranked 12th for the overall F-score and 8th for the F-score of non-regulation events.

Original language	English
Title of host publication	Proceedings of the Workshop on Current Trends in Biomedical Natural Language Processing: Shared Task
Place of Publication	Stroudsburg, PA, USA
Publisher	Association for Computational Linguistics
Pages	115-118
Number of pages	4
ISBN (Print)	978-1-932432-44-2
DOIs	https://doi.org/10.3115/1572340.1572359
Publication status	Published - 2009
Event	BioNLP '09 Proceedings of the Workshop on Current Trends in Biomedical Natural Language Processing: Shared Task - Duration: 1 Jan 1824 → …

Conference

Conference	BioNLP '09 Proceedings of the Workshop on Current Trends in Biomedical Natural Language Processing: Shared Task
Abbreviated title	BioNLP '09
Period	1/01/24 → …

Access to Document

10.3115/1572340.1572359

FULL-TEXT

Cite this

Sarafraz, F., Eales, J., Mohammadi, R., Dickerson, J., Robertson, D., & Nenadic, G. (2009). Biomedical event detection using rules, conditional random fields and parse tree distances. In Proceedings of the Workshop on Current Trends in Biomedical Natural Language Processing: Shared Task (pp. 115-118). Association for Computational Linguistics. https://doi.org/10.3115/1572340.1572359

@inproceedings{040181384d0148afbec758ca9909f1c0,

title = "Biomedical event detection using rules, conditional random fields and parse tree distances",

abstract = "This paper reports on a system developed for the BioNLP'09 shared task on detection and characterisation of biomedical events. Event triggers and types were recognised using a conditional random field classifier and a set of rules, while event participants were identified using a rule-based system that relied on relative distances between candidate entities and the trigger in the associated parse tree. The results on previously unseen test data were encouraging: for non-regulatory events, the F-score was almost 50% (with precision above 60%), with the overall F-score of around 30% (49% precision). The performance on more complex regulatory events was poor (F-measure of 7%). Among the 24 teams submitting the test results, our results were ranked 12th for the overall F-score and 8th for the F-score of non-regulation events.",

author = "Farzaneh Sarafraz and James Eales and R Mohammadi and Jonathan Dickerson and David Robertson and Goran Nenadic",

year = "2009",

doi = "10.3115/1572340.1572359",

language = "English",

isbn = "978-1-932432-44-2",

pages = "115--118",

booktitle = "Proceedings of the Workshop on Current Trends in Biomedical Natural Language Processing: Shared Task",

publisher = "Association for Computational Linguistics",

address = "United States",

note = "BioNLP '09 Proceedings of the Workshop on Current Trends in Biomedical Natural Language Processing: Shared Task, BioNLP '09 ; Conference date: 01-01-1824",

}

Sarafraz, F, Eales, J, Mohammadi, R, Dickerson, J, Robertson, D & Nenadic, G 2009, Biomedical event detection using rules, conditional random fields and parse tree distances. in Proceedings of the Workshop on Current Trends in Biomedical Natural Language Processing: Shared Task. Association for Computational Linguistics, Stroudsburg, PA, USA, pp. 115-118, BioNLP '09 Proceedings of the Workshop on Current Trends in Biomedical Natural Language Processing: Shared Task, 1/01/24. https://doi.org/10.3115/1572340.1572359

Biomedical event detection using rules, conditional random fields and parse tree distances. / Sarafraz, Farzaneh; Eales, James; Mohammadi, R et al.
Proceedings of the Workshop on Current Trends in Biomedical Natural Language Processing: Shared Task. Stroudsburg, PA, USA: Association for Computational Linguistics, 2009. p. 115-118.

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Biomedical event detection using rules, conditional random fields and parse tree distances

AU - Sarafraz, Farzaneh

AU - Eales, James

AU - Mohammadi, R

AU - Dickerson, Jonathan

AU - Robertson, David

AU - Nenadic, Goran

PY - 2009

Y1 - 2009

N2 - This paper reports on a system developed for the BioNLP'09 shared task on detection and characterisation of biomedical events. Event triggers and types were recognised using a conditional random field classifier and a set of rules, while event participants were identified using a rule-based system that relied on relative distances between candidate entities and the trigger in the associated parse tree. The results on previously unseen test data were encouraging: for non-regulatory events, the F-score was almost 50% (with precision above 60%), with the overall F-score of around 30% (49% precision). The performance on more complex regulatory events was poor (F-measure of 7%). Among the 24 teams submitting the test results, our results were ranked 12th for the overall F-score and 8th for the F-score of non-regulation events.

AB - This paper reports on a system developed for the BioNLP'09 shared task on detection and characterisation of biomedical events. Event triggers and types were recognised using a conditional random field classifier and a set of rules, while event participants were identified using a rule-based system that relied on relative distances between candidate entities and the trigger in the associated parse tree. The results on previously unseen test data were encouraging: for non-regulatory events, the F-score was almost 50% (with precision above 60%), with the overall F-score of around 30% (49% precision). The performance on more complex regulatory events was poor (F-measure of 7%). Among the 24 teams submitting the test results, our results were ranked 12th for the overall F-score and 8th for the F-score of non-regulation events.

U2 - 10.3115/1572340.1572359

DO - 10.3115/1572340.1572359

M3 - Conference contribution

SN - 978-1-932432-44-2

SP - 115

EP - 118

BT - Proceedings of the Workshop on Current Trends in Biomedical Natural Language Processing: Shared Task

PB - Association for Computational Linguistics

CY - Stroudsburg, PA, USA

T2 - BioNLP '09 Proceedings of the Workshop on Current Trends in Biomedical Natural Language Processing: Shared Task

Y2 - 1 January 1824

ER -

Sarafraz F, Eales J, Mohammadi R, Dickerson J, Robertson D, Nenadic G. Biomedical event detection using rules, conditional random fields and parse tree distances. In Proceedings of the Workshop on Current Trends in Biomedical Natural Language Processing: Shared Task. Stroudsburg, PA, USA: Association for Computational Linguistics. 2009. p. 115-118 doi: 10.3115/1572340.1572359

Biomedical event detection using rules, conditional random fields and parse tree distances

Abstract

Conference

Access to Document

Fingerprint

Cite this