Extracting chemical-protein relations with ensembles of SVM and deep learning models

Yifan Peng; Anthony Rios; Ramakanth Kavuluru; Zhiyong Lu

doi:10.1093/database/bay073

Extracting chemical-protein relations with ensembles of SVM and deep learning models

Yifan Peng, Anthony Rios, Ramakanth Kavuluru, Zhiyong Lu

Internal Medicine

Research output: Contribution to journal › Article › peer-review

55 Scopus citations

Abstract

Mining relations between chemicals and proteins from the biomedical literature is an increasingly important task. The CHEMPROT track at BioCreative VI aims to promote the development and evaluation of systems that can automatically detect the chemical- protein relations in running text (PubMed abstracts). This work describes our CHEMPROT track entry, which is an ensemble of three systems, including a support vector machine, a convolutional neural network, and a recurrent neural network. Their output is combined using majority voting or stacking for final predictions. Our CHEMPROT system obtained 0.7266 in precision and 0.5735 in recall for an F-score of 0.6410 during the challenge, demonstrating the effectiveness of machine learning-based approaches for automatic relation extraction from biomedical literature and achieving the highest performance in the task during the 2017 challenge.

Original language	English
Article number	bay073
Journal	Database
Volume	2018
Issue number	2018
DOIs	https://doi.org/10.1093/database/bay073
State	Published - Jan 1 2018

Bibliographical note

Funding Information:
This work was supported by the Intramural Research Programs of the National Institutes of Health, National Library of Medicine [Y.P. and Z.L.] and through an extramural National Library of Medicine grant [R21LM012274 to A.R. and R.K.]. A.R. was a summer intern at the NCBI/NIH and supported by the NIH Intramural Research Training Award.

Publisher Copyright:
© 2018 Oxford University Press. All rights reserved.

ASJC Scopus subject areas

Information Systems
Biochemistry, Genetics and Molecular Biology (all)
Agricultural and Biological Sciences (all)

Access to Document

10.1093/database/bay073

Cite this

@article{22c312c1f3a344d8960653206750df86,

title = "Extracting chemical-protein relations with ensembles of SVM and deep learning models",

abstract = "Mining relations between chemicals and proteins from the biomedical literature is an increasingly important task. The CHEMPROT track at BioCreative VI aims to promote the development and evaluation of systems that can automatically detect the chemical- protein relations in running text (PubMed abstracts). This work describes our CHEMPROT track entry, which is an ensemble of three systems, including a support vector machine, a convolutional neural network, and a recurrent neural network. Their output is combined using majority voting or stacking for final predictions. Our CHEMPROT system obtained 0.7266 in precision and 0.5735 in recall for an F-score of 0.6410 during the challenge, demonstrating the effectiveness of machine learning-based approaches for automatic relation extraction from biomedical literature and achieving the highest performance in the task during the 2017 challenge.",

author = "Yifan Peng and Anthony Rios and Ramakanth Kavuluru and Zhiyong Lu",

note = "Funding Information: This work was supported by the Intramural Research Programs of the National Institutes of Health, National Library of Medicine [Y.P. and Z.L.] and through an extramural National Library of Medicine grant [R21LM012274 to A.R. and R.K.]. A.R. was a summer intern at the NCBI/NIH and supported by the NIH Intramural Research Training Award. Publisher Copyright: {\textcopyright} 2018 Oxford University Press. All rights reserved.",

year = "2018",

month = jan,

day = "1",

doi = "10.1093/database/bay073",

language = "English",

volume = "2018",

number = "2018",

}

TY - JOUR

T1 - Extracting chemical-protein relations with ensembles of SVM and deep learning models

AU - Peng, Yifan

AU - Rios, Anthony

AU - Kavuluru, Ramakanth

AU - Lu, Zhiyong

N1 - Funding Information: This work was supported by the Intramural Research Programs of the National Institutes of Health, National Library of Medicine [Y.P. and Z.L.] and through an extramural National Library of Medicine grant [R21LM012274 to A.R. and R.K.]. A.R. was a summer intern at the NCBI/NIH and supported by the NIH Intramural Research Training Award. Publisher Copyright: © 2018 Oxford University Press. All rights reserved.

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Mining relations between chemicals and proteins from the biomedical literature is an increasingly important task. The CHEMPROT track at BioCreative VI aims to promote the development and evaluation of systems that can automatically detect the chemical- protein relations in running text (PubMed abstracts). This work describes our CHEMPROT track entry, which is an ensemble of three systems, including a support vector machine, a convolutional neural network, and a recurrent neural network. Their output is combined using majority voting or stacking for final predictions. Our CHEMPROT system obtained 0.7266 in precision and 0.5735 in recall for an F-score of 0.6410 during the challenge, demonstrating the effectiveness of machine learning-based approaches for automatic relation extraction from biomedical literature and achieving the highest performance in the task during the 2017 challenge.

AB - Mining relations between chemicals and proteins from the biomedical literature is an increasingly important task. The CHEMPROT track at BioCreative VI aims to promote the development and evaluation of systems that can automatically detect the chemical- protein relations in running text (PubMed abstracts). This work describes our CHEMPROT track entry, which is an ensemble of three systems, including a support vector machine, a convolutional neural network, and a recurrent neural network. Their output is combined using majority voting or stacking for final predictions. Our CHEMPROT system obtained 0.7266 in precision and 0.5735 in recall for an F-score of 0.6410 during the challenge, demonstrating the effectiveness of machine learning-based approaches for automatic relation extraction from biomedical literature and achieving the highest performance in the task during the 2017 challenge.

UR - http://www.scopus.com/inward/record.url?scp=85056073558&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85056073558&partnerID=8YFLogxK

U2 - 10.1093/database/bay073

DO - 10.1093/database/bay073

M3 - Article

C2 - 30020437

AN - SCOPUS:85056073558

SN - 1758-0463

VL - 2018

JO - Database

JF - Database

IS - 2018

M1 - bay073

ER -

Extracting chemical-protein relations with ensembles of SVM and deep learning models

Abstract

Bibliographical note

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this