Grounded feature selection for biomedical relation extraction by the combinative approach

Sung Jeon Song, Go Eun Heo, Ha Jin Kim, Hyo Jung Jung, Yong Hwan Kim, Min Song

Research output: Chapter in Book/Report/Conference proceedingConference contribution

7 Citations (Scopus)

Abstract

Relation extraction is an important task in biomedical areas such as protein-protein interaction, gene-disease interactions, and drugdisease interactions. In recent years, it has been widely researched to automatically extract biomedical relations in a vest amount of biomedical text data. In this paper, we propose a hybrid approach to extracting relations based on a rule-based approach feature set. We then use different classification algorithms such as SVM, Naïve Bayes, and Decision Tree classifiers for relation classification. The rationale for adopting shallow parsing and other NLP techniques to extract relations is two-folds: simplicity and robustness. We select seven features with the rule-based shallow parsing technique and evaluate the performance with four different PPI public corpora. Our experimental results show the stable performance in F-measure even with the relatively fewer features.

Original languageEnglish
Title of host publicationDTMBIO 2014 - Proceedings of the ACM 8th International Workshop on Data and Text Mining in Bioinformatics, co-located with CIKM 2014
PublisherAssociation for Computing Machinery, Inc
Pages29-32
Number of pages4
ISBN (Electronic)9781450312752
DOIs
Publication statusPublished - 2014 Nov 7
Event8th ACM International Workshop on Data and Text Mining in Biomedical Informatics, DTMBIO 2014 - Shanghai, China
Duration: 2014 Nov 7 → …

Other

Other8th ACM International Workshop on Data and Text Mining in Biomedical Informatics, DTMBIO 2014
CountryChina
CityShanghai
Period14/11/7 → …

Fingerprint

Feature extraction
Proteins
Decision trees
Classifiers
Genes

All Science Journal Classification (ASJC) codes

  • Computational Theory and Mathematics
  • Computer Science Applications

Cite this

Song, S. J., Heo, G. E., Kim, H. J., Jung, H. J., Kim, Y. H., & Song, M. (2014). Grounded feature selection for biomedical relation extraction by the combinative approach. In DTMBIO 2014 - Proceedings of the ACM 8th International Workshop on Data and Text Mining in Bioinformatics, co-located with CIKM 2014 (pp. 29-32). Association for Computing Machinery, Inc. https://doi.org/10.1145/2665970.2665975
Song, Sung Jeon ; Heo, Go Eun ; Kim, Ha Jin ; Jung, Hyo Jung ; Kim, Yong Hwan ; Song, Min. / Grounded feature selection for biomedical relation extraction by the combinative approach. DTMBIO 2014 - Proceedings of the ACM 8th International Workshop on Data and Text Mining in Bioinformatics, co-located with CIKM 2014. Association for Computing Machinery, Inc, 2014. pp. 29-32
@inproceedings{ada3416b53a1496aab0d5962afb6a1fe,
title = "Grounded feature selection for biomedical relation extraction by the combinative approach",
abstract = "Relation extraction is an important task in biomedical areas such as protein-protein interaction, gene-disease interactions, and drugdisease interactions. In recent years, it has been widely researched to automatically extract biomedical relations in a vest amount of biomedical text data. In this paper, we propose a hybrid approach to extracting relations based on a rule-based approach feature set. We then use different classification algorithms such as SVM, Na{\"i}ve Bayes, and Decision Tree classifiers for relation classification. The rationale for adopting shallow parsing and other NLP techniques to extract relations is two-folds: simplicity and robustness. We select seven features with the rule-based shallow parsing technique and evaluate the performance with four different PPI public corpora. Our experimental results show the stable performance in F-measure even with the relatively fewer features.",
author = "Song, {Sung Jeon} and Heo, {Go Eun} and Kim, {Ha Jin} and Jung, {Hyo Jung} and Kim, {Yong Hwan} and Min Song",
year = "2014",
month = "11",
day = "7",
doi = "10.1145/2665970.2665975",
language = "English",
pages = "29--32",
booktitle = "DTMBIO 2014 - Proceedings of the ACM 8th International Workshop on Data and Text Mining in Bioinformatics, co-located with CIKM 2014",
publisher = "Association for Computing Machinery, Inc",

}

Song, SJ, Heo, GE, Kim, HJ, Jung, HJ, Kim, YH & Song, M 2014, Grounded feature selection for biomedical relation extraction by the combinative approach. in DTMBIO 2014 - Proceedings of the ACM 8th International Workshop on Data and Text Mining in Bioinformatics, co-located with CIKM 2014. Association for Computing Machinery, Inc, pp. 29-32, 8th ACM International Workshop on Data and Text Mining in Biomedical Informatics, DTMBIO 2014, Shanghai, China, 14/11/7. https://doi.org/10.1145/2665970.2665975

Grounded feature selection for biomedical relation extraction by the combinative approach. / Song, Sung Jeon; Heo, Go Eun; Kim, Ha Jin; Jung, Hyo Jung; Kim, Yong Hwan; Song, Min.

DTMBIO 2014 - Proceedings of the ACM 8th International Workshop on Data and Text Mining in Bioinformatics, co-located with CIKM 2014. Association for Computing Machinery, Inc, 2014. p. 29-32.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Grounded feature selection for biomedical relation extraction by the combinative approach

AU - Song, Sung Jeon

AU - Heo, Go Eun

AU - Kim, Ha Jin

AU - Jung, Hyo Jung

AU - Kim, Yong Hwan

AU - Song, Min

PY - 2014/11/7

Y1 - 2014/11/7

N2 - Relation extraction is an important task in biomedical areas such as protein-protein interaction, gene-disease interactions, and drugdisease interactions. In recent years, it has been widely researched to automatically extract biomedical relations in a vest amount of biomedical text data. In this paper, we propose a hybrid approach to extracting relations based on a rule-based approach feature set. We then use different classification algorithms such as SVM, Naïve Bayes, and Decision Tree classifiers for relation classification. The rationale for adopting shallow parsing and other NLP techniques to extract relations is two-folds: simplicity and robustness. We select seven features with the rule-based shallow parsing technique and evaluate the performance with four different PPI public corpora. Our experimental results show the stable performance in F-measure even with the relatively fewer features.

AB - Relation extraction is an important task in biomedical areas such as protein-protein interaction, gene-disease interactions, and drugdisease interactions. In recent years, it has been widely researched to automatically extract biomedical relations in a vest amount of biomedical text data. In this paper, we propose a hybrid approach to extracting relations based on a rule-based approach feature set. We then use different classification algorithms such as SVM, Naïve Bayes, and Decision Tree classifiers for relation classification. The rationale for adopting shallow parsing and other NLP techniques to extract relations is two-folds: simplicity and robustness. We select seven features with the rule-based shallow parsing technique and evaluate the performance with four different PPI public corpora. Our experimental results show the stable performance in F-measure even with the relatively fewer features.

UR - http://www.scopus.com/inward/record.url?scp=84974602511&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84974602511&partnerID=8YFLogxK

U2 - 10.1145/2665970.2665975

DO - 10.1145/2665970.2665975

M3 - Conference contribution

SP - 29

EP - 32

BT - DTMBIO 2014 - Proceedings of the ACM 8th International Workshop on Data and Text Mining in Bioinformatics, co-located with CIKM 2014

PB - Association for Computing Machinery, Inc

ER -

Song SJ, Heo GE, Kim HJ, Jung HJ, Kim YH, Song M. Grounded feature selection for biomedical relation extraction by the combinative approach. In DTMBIO 2014 - Proceedings of the ACM 8th International Workshop on Data and Text Mining in Bioinformatics, co-located with CIKM 2014. Association for Computing Machinery, Inc. 2014. p. 29-32 https://doi.org/10.1145/2665970.2665975