From actors, politicians, to CEOs: Domain adaptation of relational extractors using a latent relational mapping

Danushka Bollegala, Yutaka Matsuo, Mitsuru Ishizuka

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

Abstract

We propose a method to adapt an existing relation extraction system to extract new relation types with minimum supervision. Our proposed method comprises two stages: learning a lower-dimensional projection between different relations, and learning a relational classifier for the target relation type with instance sampling. We evaluate the proposed method using a dataset that contains 2000 instances for 20 different relation types. Our experimental results show that the proposed method achieves a statistically significant macro-average F-score of 62.77. Moreover, the proposed method outperforms numerous baselines and a previously proposed weakly-supervised relation extraction method.

Original languageEnglish
Title of host publicationProceedings of the 20th International Conference Companion on World Wide Web, WWW 2011
Pages13-14
Number of pages2
DOIs
Publication statusPublished - 2011
Externally publishedYes
Event20th International Conference Companion on World Wide Web, WWW 2011 - Hyderabad
Duration: 2011 Mar 282011 Apr 1

Other

Other20th International Conference Companion on World Wide Web, WWW 2011
CityHyderabad
Period11/3/2811/4/1

Fingerprint

Macros
Classifiers
Sampling

Keywords

  • domain adaptation
  • relation extraction
  • web mining

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems

Cite this

Bollegala, D., Matsuo, Y., & Ishizuka, M. (2011). From actors, politicians, to CEOs: Domain adaptation of relational extractors using a latent relational mapping. In Proceedings of the 20th International Conference Companion on World Wide Web, WWW 2011 (pp. 13-14) https://doi.org/10.1145/1963192.1963200

From actors, politicians, to CEOs : Domain adaptation of relational extractors using a latent relational mapping. / Bollegala, Danushka; Matsuo, Yutaka; Ishizuka, Mitsuru.

Proceedings of the 20th International Conference Companion on World Wide Web, WWW 2011. 2011. p. 13-14.

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

Bollegala, D, Matsuo, Y & Ishizuka, M 2011, From actors, politicians, to CEOs: Domain adaptation of relational extractors using a latent relational mapping. in Proceedings of the 20th International Conference Companion on World Wide Web, WWW 2011. pp. 13-14, 20th International Conference Companion on World Wide Web, WWW 2011, Hyderabad, 11/3/28. https://doi.org/10.1145/1963192.1963200
Bollegala D, Matsuo Y, Ishizuka M. From actors, politicians, to CEOs: Domain adaptation of relational extractors using a latent relational mapping. In Proceedings of the 20th International Conference Companion on World Wide Web, WWW 2011. 2011. p. 13-14 https://doi.org/10.1145/1963192.1963200
Bollegala, Danushka ; Matsuo, Yutaka ; Ishizuka, Mitsuru. / From actors, politicians, to CEOs : Domain adaptation of relational extractors using a latent relational mapping. Proceedings of the 20th International Conference Companion on World Wide Web, WWW 2011. 2011. pp. 13-14
@inproceedings{46908cb5ced840f5935056d02ac11609,
title = "From actors, politicians, to CEOs: Domain adaptation of relational extractors using a latent relational mapping",
abstract = "We propose a method to adapt an existing relation extraction system to extract new relation types with minimum supervision. Our proposed method comprises two stages: learning a lower-dimensional projection between different relations, and learning a relational classifier for the target relation type with instance sampling. We evaluate the proposed method using a dataset that contains 2000 instances for 20 different relation types. Our experimental results show that the proposed method achieves a statistically significant macro-average F-score of 62.77. Moreover, the proposed method outperforms numerous baselines and a previously proposed weakly-supervised relation extraction method.",
keywords = "domain adaptation, relation extraction, web mining",
author = "Danushka Bollegala and Yutaka Matsuo and Mitsuru Ishizuka",
year = "2011",
doi = "10.1145/1963192.1963200",
language = "English",
isbn = "9781450305181",
pages = "13--14",
booktitle = "Proceedings of the 20th International Conference Companion on World Wide Web, WWW 2011",

}

TY - GEN

T1 - From actors, politicians, to CEOs

T2 - Domain adaptation of relational extractors using a latent relational mapping

AU - Bollegala, Danushka

AU - Matsuo, Yutaka

AU - Ishizuka, Mitsuru

PY - 2011

Y1 - 2011

N2 - We propose a method to adapt an existing relation extraction system to extract new relation types with minimum supervision. Our proposed method comprises two stages: learning a lower-dimensional projection between different relations, and learning a relational classifier for the target relation type with instance sampling. We evaluate the proposed method using a dataset that contains 2000 instances for 20 different relation types. Our experimental results show that the proposed method achieves a statistically significant macro-average F-score of 62.77. Moreover, the proposed method outperforms numerous baselines and a previously proposed weakly-supervised relation extraction method.

AB - We propose a method to adapt an existing relation extraction system to extract new relation types with minimum supervision. Our proposed method comprises two stages: learning a lower-dimensional projection between different relations, and learning a relational classifier for the target relation type with instance sampling. We evaluate the proposed method using a dataset that contains 2000 instances for 20 different relation types. Our experimental results show that the proposed method achieves a statistically significant macro-average F-score of 62.77. Moreover, the proposed method outperforms numerous baselines and a previously proposed weakly-supervised relation extraction method.

KW - domain adaptation

KW - relation extraction

KW - web mining

UR - http://www.scopus.com/inward/record.url?scp=79955159354&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79955159354&partnerID=8YFLogxK

U2 - 10.1145/1963192.1963200

DO - 10.1145/1963192.1963200

M3 - Conference contribution

AN - SCOPUS:79955159354

SN - 9781450305181

SP - 13

EP - 14

BT - Proceedings of the 20th International Conference Companion on World Wide Web, WWW 2011

ER -