Infusion of labeled data into distant supervision for relation extraction

Maria Pershina, Bonan Min, Wei Xu, Ralph Grishman

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

Abstract

Distant supervision usually utilizes only unlabeled data and existing knowledge bases to learn relation extraction models. However, in some cases a small amount of human labeled data is available. In this paper, we demonstrate how a state-of-theart multi-instance multi-label model can be modified to make use of these reliable sentence-level labels in addition to the relation-level distant supervision from a database. Experiments show that our approach achieves a statistically significant increase of 13.5% in F-score and 37% in area under the precision recall curve.

Original languageEnglish (US)
Title of host publicationLong Papers
PublisherAssociation for Computational Linguistics (ACL)
Pages732-738
Number of pages7
Volume2
ISBN (Print)9781937284732
StatePublished - 2014
Event52nd Annual Meeting of the Association for Computational Linguistics, ACL 2014 - Baltimore, MD, United States
Duration: Jun 22 2014Jun 27 2014

Other

Other52nd Annual Meeting of the Association for Computational Linguistics, ACL 2014
CountryUnited States
CityBaltimore, MD
Period6/22/146/27/14

Fingerprint

supervision
experiment
knowledge
Supervision
Data Base
Experiment

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language

Cite this

Pershina, M., Min, B., Xu, W., & Grishman, R. (2014). Infusion of labeled data into distant supervision for relation extraction. In Long Papers (Vol. 2, pp. 732-738). Association for Computational Linguistics (ACL).

Infusion of labeled data into distant supervision for relation extraction. / Pershina, Maria; Min, Bonan; Xu, Wei; Grishman, Ralph.

Long Papers. Vol. 2 Association for Computational Linguistics (ACL), 2014. p. 732-738.

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

Pershina, M, Min, B, Xu, W & Grishman, R 2014, Infusion of labeled data into distant supervision for relation extraction. in Long Papers. vol. 2, Association for Computational Linguistics (ACL), pp. 732-738, 52nd Annual Meeting of the Association for Computational Linguistics, ACL 2014, Baltimore, MD, United States, 6/22/14.
Pershina M, Min B, Xu W, Grishman R. Infusion of labeled data into distant supervision for relation extraction. In Long Papers. Vol. 2. Association for Computational Linguistics (ACL). 2014. p. 732-738
Pershina, Maria ; Min, Bonan ; Xu, Wei ; Grishman, Ralph. / Infusion of labeled data into distant supervision for relation extraction. Long Papers. Vol. 2 Association for Computational Linguistics (ACL), 2014. pp. 732-738
@inproceedings{a9fb0a859ac141b780e2e20dbd211101,
title = "Infusion of labeled data into distant supervision for relation extraction",
abstract = "Distant supervision usually utilizes only unlabeled data and existing knowledge bases to learn relation extraction models. However, in some cases a small amount of human labeled data is available. In this paper, we demonstrate how a state-of-theart multi-instance multi-label model can be modified to make use of these reliable sentence-level labels in addition to the relation-level distant supervision from a database. Experiments show that our approach achieves a statistically significant increase of 13.5{\%} in F-score and 37{\%} in area under the precision recall curve.",
author = "Maria Pershina and Bonan Min and Wei Xu and Ralph Grishman",
year = "2014",
language = "English (US)",
isbn = "9781937284732",
volume = "2",
pages = "732--738",
booktitle = "Long Papers",
publisher = "Association for Computational Linguistics (ACL)",

}

TY - GEN

T1 - Infusion of labeled data into distant supervision for relation extraction

AU - Pershina, Maria

AU - Min, Bonan

AU - Xu, Wei

AU - Grishman, Ralph

PY - 2014

Y1 - 2014

N2 - Distant supervision usually utilizes only unlabeled data and existing knowledge bases to learn relation extraction models. However, in some cases a small amount of human labeled data is available. In this paper, we demonstrate how a state-of-theart multi-instance multi-label model can be modified to make use of these reliable sentence-level labels in addition to the relation-level distant supervision from a database. Experiments show that our approach achieves a statistically significant increase of 13.5% in F-score and 37% in area under the precision recall curve.

AB - Distant supervision usually utilizes only unlabeled data and existing knowledge bases to learn relation extraction models. However, in some cases a small amount of human labeled data is available. In this paper, we demonstrate how a state-of-theart multi-instance multi-label model can be modified to make use of these reliable sentence-level labels in addition to the relation-level distant supervision from a database. Experiments show that our approach achieves a statistically significant increase of 13.5% in F-score and 37% in area under the precision recall curve.

UR - http://www.scopus.com/inward/record.url?scp=84906924687&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84906924687&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84906924687

SN - 9781937284732

VL - 2

SP - 732

EP - 738

BT - Long Papers

PB - Association for Computational Linguistics (ACL)

ER -