A maximum entropy-based bio-molecular event extraction model that considers event generation

Hyoung Gyu Lee, So Young Park, Hae-Chang Rim, Do Gil Lee, Hong Woo Chun

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

In this paper, we propose a maximum entropy-based model, which can mathematically explain the biomolecular event extraction problem. The proposed model generates an event table, which can represent the relationship between an event trigger and its arguments. The complex sentences with distinctive event structures can be also represented by the event table. Previous approaches intuitively designed a pipeline system, which sequentially performs trigger detection and arguments recognition, and thus, did not clearly explain the relationship between identified triggers and arguments. On the other hand, the proposed model generates an event table that can represent triggers, their arguments, and their relationships. The desired events can be easily extracted from the event table. Experimental results show that the proposed model can cover 91.36% of events in the training dataset and that it can achieve a 50.44% recall in the test dataset by using the event table.

Original languageEnglish
Pages (from-to)248-265
Number of pages18
JournalJournal of Information Processing Systems
Volume11
Issue number2
DOIs
Publication statusPublished - 2014

Fingerprint

Entropy
Pipelines

Keywords

  • Bioinformatics
  • Event extraction
  • Maximum entropy
  • Text-mining

ASJC Scopus subject areas

  • Information Systems
  • Software

Cite this

A maximum entropy-based bio-molecular event extraction model that considers event generation. / Lee, Hyoung Gyu; Park, So Young; Rim, Hae-Chang; Lee, Do Gil; Chun, Hong Woo.

In: Journal of Information Processing Systems, Vol. 11, No. 2, 2014, p. 248-265.

Research output: Contribution to journalArticle

Lee, Hyoung Gyu ; Park, So Young ; Rim, Hae-Chang ; Lee, Do Gil ; Chun, Hong Woo. / A maximum entropy-based bio-molecular event extraction model that considers event generation. In: Journal of Information Processing Systems. 2014 ; Vol. 11, No. 2. pp. 248-265.
@article{d3ed1fc42cc4411abdcd321a6e05c1e8,
title = "A maximum entropy-based bio-molecular event extraction model that considers event generation",
abstract = "In this paper, we propose a maximum entropy-based model, which can mathematically explain the biomolecular event extraction problem. The proposed model generates an event table, which can represent the relationship between an event trigger and its arguments. The complex sentences with distinctive event structures can be also represented by the event table. Previous approaches intuitively designed a pipeline system, which sequentially performs trigger detection and arguments recognition, and thus, did not clearly explain the relationship between identified triggers and arguments. On the other hand, the proposed model generates an event table that can represent triggers, their arguments, and their relationships. The desired events can be easily extracted from the event table. Experimental results show that the proposed model can cover 91.36{\%} of events in the training dataset and that it can achieve a 50.44{\%} recall in the test dataset by using the event table.",
keywords = "Bioinformatics, Event extraction, Maximum entropy, Text-mining",
author = "Lee, {Hyoung Gyu} and Park, {So Young} and Hae-Chang Rim and Lee, {Do Gil} and Chun, {Hong Woo}",
year = "2014",
doi = "10.3745/JIPS.04.0008",
language = "English",
volume = "11",
pages = "248--265",
journal = "Journal of Information Processing Systems",
issn = "1976-913X",
publisher = "Korea Information Processing Society (KIPS)",
number = "2",

}

TY - JOUR

T1 - A maximum entropy-based bio-molecular event extraction model that considers event generation

AU - Lee, Hyoung Gyu

AU - Park, So Young

AU - Rim, Hae-Chang

AU - Lee, Do Gil

AU - Chun, Hong Woo

PY - 2014

Y1 - 2014

N2 - In this paper, we propose a maximum entropy-based model, which can mathematically explain the biomolecular event extraction problem. The proposed model generates an event table, which can represent the relationship between an event trigger and its arguments. The complex sentences with distinctive event structures can be also represented by the event table. Previous approaches intuitively designed a pipeline system, which sequentially performs trigger detection and arguments recognition, and thus, did not clearly explain the relationship between identified triggers and arguments. On the other hand, the proposed model generates an event table that can represent triggers, their arguments, and their relationships. The desired events can be easily extracted from the event table. Experimental results show that the proposed model can cover 91.36% of events in the training dataset and that it can achieve a 50.44% recall in the test dataset by using the event table.

AB - In this paper, we propose a maximum entropy-based model, which can mathematically explain the biomolecular event extraction problem. The proposed model generates an event table, which can represent the relationship between an event trigger and its arguments. The complex sentences with distinctive event structures can be also represented by the event table. Previous approaches intuitively designed a pipeline system, which sequentially performs trigger detection and arguments recognition, and thus, did not clearly explain the relationship between identified triggers and arguments. On the other hand, the proposed model generates an event table that can represent triggers, their arguments, and their relationships. The desired events can be easily extracted from the event table. Experimental results show that the proposed model can cover 91.36% of events in the training dataset and that it can achieve a 50.44% recall in the test dataset by using the event table.

KW - Bioinformatics

KW - Event extraction

KW - Maximum entropy

KW - Text-mining

UR - http://www.scopus.com/inward/record.url?scp=84945120678&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84945120678&partnerID=8YFLogxK

U2 - 10.3745/JIPS.04.0008

DO - 10.3745/JIPS.04.0008

M3 - Article

AN - SCOPUS:84945120678

VL - 11

SP - 248

EP - 265

JO - Journal of Information Processing Systems

JF - Journal of Information Processing Systems

SN - 1976-913X

IS - 2

ER -