Multi-source named entity typing for social media

Reuth Vexler; Einat Minkov

doi:10.18653/v1/W16-2702

Multi-source named entity typing for social media

Reuth Vexler, Einat Minkov

Department of Information Systems

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Typed lexicons that encode knowledge about the semantic types of an entity
name, e.g., that ‘Paris’ denotes a geolocation, product, or person, have proven useful for many text processing tasks. While lexicons may be derived from large-scale knowledge bases (KBs), KBs are inherently imperfect, in particular they lack
coverage with respect to long tail entity names. We infer the types of a given
entity name using multi-source learning, considering information obtained by
alignment to the Freebase knowledge base, Web-scale distributional patterns,
and global semi-structured contexts retrieved by means of Web search. Evaluation in the challenging domain of social media shows that multi-source learning improves performance compared with rule-based KB lookups, boosting typing results for some semantic categories.

Original language	English
Title of host publication	Proceedings of NEWS 2016
Subtitle of host publication	6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016
Editors	Xiangyu Duan, Rafael E. Banchs, Min Zhang, Haizhou Li, A. Kumara
Place of Publication	Berlin, Germany
Publisher	Association for Computational Linguistics (ACL)
Pages	11-20
Number of pages	10
ISBN (Electronic)	9781945626166
DOIs	https://doi.org/10.18653/v1/W16-2702
State	Published - 1 Aug 2016
Event	6th Named Entity Workshop, NEWS 2016 at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016 - Berlin, Germany Duration: 12 Aug 2016 → …

Publication series

Name	Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016

Conference

Conference	6th Named Entity Workshop, NEWS 2016 at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016
Country/Territory	Germany
City	Berlin
Period	12/08/16 → …

Bibliographical note

Publisher Copyright:
© Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016. All rights reserved.

ASJC Scopus subject areas

Computer Science Applications
Computational Theory and Mathematics
Software

Access to Document

10.18653/v1/W16-2702

https://aclanthology.org/W16-2702

Cite this

Vexler, R., & Minkov, E. (2016). Multi-source named entity typing for social media. In X. Duan, R. E. Banchs, M. Zhang, H. Li, & A. Kumara (Eds.), Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016 (pp. 11-20). (Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/W16-2702

Vexler, Reuth ; Minkov, Einat. / Multi-source named entity typing for social media. Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016. editor / Xiangyu Duan ; Rafael E. Banchs ; Min Zhang ; Haizhou Li ; A. Kumara. Berlin, Germany : Association for Computational Linguistics (ACL), 2016. pp. 11-20 (Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016).

@inproceedings{c6b1d68c2deb4e6d92b553b729fba5a0,

title = "Multi-source named entity typing for social media",

abstract = "Typed lexicons that encode knowledge about the semantic types of an entityname, e.g., that {\textquoteleft}Paris{\textquoteright} denotes a geolocation, product, or person, have proven useful for many text processing tasks. While lexicons may be derived from large-scale knowledge bases (KBs), KBs are inherently imperfect, in particular they lackcoverage with respect to long tail entity names. We infer the types of a givenentity name using multi-source learning, considering information obtained byalignment to the Freebase knowledge base, Web-scale distributional patterns,and global semi-structured contexts retrieved by means of Web search. Evaluation in the challenging domain of social media shows that multi-source learning improves performance compared with rule-based KB lookups, boosting typing results for some semantic categories.",

author = "Reuth Vexler and Einat Minkov",

note = "Publisher Copyright: {\textcopyright} Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016. All rights reserved.; 6th Named Entity Workshop, NEWS 2016 at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016 ; Conference date: 12-08-2016",

year = "2016",

month = aug,

day = "1",

doi = "10.18653/v1/W16-2702",

language = "אנגלית",

series = "Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016",

publisher = "Association for Computational Linguistics (ACL)",

pages = "11--20",

editor = "Xiangyu Duan and Banchs, {Rafael E.} and Min Zhang and Haizhou Li and A. Kumara",

booktitle = "Proceedings of NEWS 2016",

address = "ארצות הברית",

}

Vexler, R & Minkov, E 2016, Multi-source named entity typing for social media. in X Duan, RE Banchs, M Zhang, H Li & A Kumara (eds), Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016. Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016, Association for Computational Linguistics (ACL), Berlin, Germany, pp. 11-20, 6th Named Entity Workshop, NEWS 2016 at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016, Berlin, Germany, 12/08/16. https://doi.org/10.18653/v1/W16-2702

Multi-source named entity typing for social media. / Vexler, Reuth; Minkov, Einat.
Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016. ed. / Xiangyu Duan; Rafael E. Banchs; Min Zhang; Haizhou Li; A. Kumara. Berlin, Germany: Association for Computational Linguistics (ACL), 2016. p. 11-20 (Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Multi-source named entity typing for social media

AU - Vexler, Reuth

AU - Minkov, Einat

PY - 2016/8/1

Y1 - 2016/8/1

N2 - Typed lexicons that encode knowledge about the semantic types of an entityname, e.g., that ‘Paris’ denotes a geolocation, product, or person, have proven useful for many text processing tasks. While lexicons may be derived from large-scale knowledge bases (KBs), KBs are inherently imperfect, in particular they lackcoverage with respect to long tail entity names. We infer the types of a givenentity name using multi-source learning, considering information obtained byalignment to the Freebase knowledge base, Web-scale distributional patterns,and global semi-structured contexts retrieved by means of Web search. Evaluation in the challenging domain of social media shows that multi-source learning improves performance compared with rule-based KB lookups, boosting typing results for some semantic categories.

AB - Typed lexicons that encode knowledge about the semantic types of an entityname, e.g., that ‘Paris’ denotes a geolocation, product, or person, have proven useful for many text processing tasks. While lexicons may be derived from large-scale knowledge bases (KBs), KBs are inherently imperfect, in particular they lackcoverage with respect to long tail entity names. We infer the types of a givenentity name using multi-source learning, considering information obtained byalignment to the Freebase knowledge base, Web-scale distributional patterns,and global semi-structured contexts retrieved by means of Web search. Evaluation in the challenging domain of social media shows that multi-source learning improves performance compared with rule-based KB lookups, boosting typing results for some semantic categories.

UR - http://www.scopus.com/inward/record.url?scp=85060730538&partnerID=8YFLogxK

U2 - 10.18653/v1/W16-2702

DO - 10.18653/v1/W16-2702

M3 - פרסום בספר כנס

T3 - Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016

SP - 11

EP - 20

BT - Proceedings of NEWS 2016

A2 - Duan, Xiangyu

A2 - Banchs, Rafael E.

A2 - Zhang, Min

A2 - Li, Haizhou

A2 - Kumara, A.

PB - Association for Computational Linguistics (ACL)

CY - Berlin, Germany

T2 - 6th Named Entity Workshop, NEWS 2016 at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016

Y2 - 12 August 2016

ER -

Vexler R, Minkov E. Multi-source named entity typing for social media. In Duan X, Banchs RE, Zhang M, Li H, Kumara A, editors, Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016. Berlin, Germany: Association for Computational Linguistics (ACL). 2016. p. 11-20. (Proceedings of NEWS 2016: 6th Named Entity Workshop at the 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016). doi: 10.18653/v1/W16-2702

Multi-source named entity typing for social media

Abstract

Publication series

Conference

Bibliographical note

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this