Improving name tagging by reference resolution and relation detection

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

Abstract

Information extraction systems incorporate multiple stages of linguistic analysis. Although errors are typically compounded from stage to stage, it is possible to reduce the errors in one stage by harnessing the results of the other stages. We demonstrate this by using the results of coreference analysis and relation extraction to reduce the errors produced by a Chinese name tagger. We use an N-best approach to generate multiple hypotheses and have them re-ranked by subsequent stages of processing. We obtained thereby a reduction of 24% in spurious and incorrect name tags, and a reduction of 14% in missed tags.

Original languageEnglish (US)
Title of host publicationACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference
Pages411-418
Number of pages8
StatePublished - 2005
Event43rd Annual Meeting of the Association for Computational Linguistics, ACL-05 - Ann Arbor, MI, United States
Duration: Jun 25 2005Jun 30 2005

Other

Other43rd Annual Meeting of the Association for Computational Linguistics, ACL-05
CountryUnited States
CityAnn Arbor, MI
Period6/25/056/30/05

Fingerprint

linguistics
Reference Resolution
Names
Tag
Tagging
Linguistic Analysis
Coreference
Information Extraction

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language

Cite this

Ji, H., & Grishman, R. (2005). Improving name tagging by reference resolution and relation detection. In ACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (pp. 411-418)

Improving name tagging by reference resolution and relation detection. / Ji, Heng; Grishman, Ralph.

ACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference. 2005. p. 411-418.

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

Ji, H & Grishman, R 2005, Improving name tagging by reference resolution and relation detection. in ACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference. pp. 411-418, 43rd Annual Meeting of the Association for Computational Linguistics, ACL-05, Ann Arbor, MI, United States, 6/25/05.
Ji H, Grishman R. Improving name tagging by reference resolution and relation detection. In ACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference. 2005. p. 411-418
Ji, Heng ; Grishman, Ralph. / Improving name tagging by reference resolution and relation detection. ACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference. 2005. pp. 411-418
@inproceedings{9875cfdb6b2346a1a01d49fae047474c,
title = "Improving name tagging by reference resolution and relation detection",
abstract = "Information extraction systems incorporate multiple stages of linguistic analysis. Although errors are typically compounded from stage to stage, it is possible to reduce the errors in one stage by harnessing the results of the other stages. We demonstrate this by using the results of coreference analysis and relation extraction to reduce the errors produced by a Chinese name tagger. We use an N-best approach to generate multiple hypotheses and have them re-ranked by subsequent stages of processing. We obtained thereby a reduction of 24{\%} in spurious and incorrect name tags, and a reduction of 14{\%} in missed tags.",
author = "Heng Ji and Ralph Grishman",
year = "2005",
language = "English (US)",
isbn = "1932432515",
pages = "411--418",
booktitle = "ACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference",

}

TY - GEN

T1 - Improving name tagging by reference resolution and relation detection

AU - Ji, Heng

AU - Grishman, Ralph

PY - 2005

Y1 - 2005

N2 - Information extraction systems incorporate multiple stages of linguistic analysis. Although errors are typically compounded from stage to stage, it is possible to reduce the errors in one stage by harnessing the results of the other stages. We demonstrate this by using the results of coreference analysis and relation extraction to reduce the errors produced by a Chinese name tagger. We use an N-best approach to generate multiple hypotheses and have them re-ranked by subsequent stages of processing. We obtained thereby a reduction of 24% in spurious and incorrect name tags, and a reduction of 14% in missed tags.

AB - Information extraction systems incorporate multiple stages of linguistic analysis. Although errors are typically compounded from stage to stage, it is possible to reduce the errors in one stage by harnessing the results of the other stages. We demonstrate this by using the results of coreference analysis and relation extraction to reduce the errors produced by a Chinese name tagger. We use an N-best approach to generate multiple hypotheses and have them re-ranked by subsequent stages of processing. We obtained thereby a reduction of 24% in spurious and incorrect name tags, and a reduction of 14% in missed tags.

UR - http://www.scopus.com/inward/record.url?scp=84859886499&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84859886499&partnerID=8YFLogxK

M3 - Conference contribution

SN - 1932432515

SN - 9781932432510

SP - 411

EP - 418

BT - ACL-05 - 43rd Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference

ER -