Refining event extraction through cross-document inference

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

Abstract

We apply the hypothesis of "One Sense Per Discourse" (Yarowsky, 1995) to information extraction (IE), and extend the scope of "discourse" from one single document to a cluster of topically-related documents. We employ a similar approach to propagate consistent event arguments across sentences and documents. Combining global evidence from related documents with local decisions, we design a simple scheme to conduct cross-document inference for improving the ACE event extraction task1. Without using any additional labeled data this new approach obtained 7.6% higher F-Measure in trigger labeling and 6% higher F-Measure in argument labeling over a state-of-the-art IE system which extracts events independently for each sentence.

Original languageEnglish (US)
Title of host publicationACL-08: HLT - 46th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference
Pages254-262
Number of pages9
StatePublished - 2008
Event46th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, ACL-08: HLT - Columbus, OH, United States
Duration: Jun 15 2008Jun 20 2008

Other

Other46th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, ACL-08: HLT
CountryUnited States
CityColumbus, OH
Period6/15/086/20/08

Fingerprint

Refining
Labeling
event
discourse
Inference
Information Extraction
Discourse
evidence
Trigger

ASJC Scopus subject areas

  • Language and Linguistics
  • Computer Networks and Communications
  • Linguistics and Language

Cite this

Ji, H., & Grishman, R. (2008). Refining event extraction through cross-document inference. In ACL-08: HLT - 46th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference (pp. 254-262)

Refining event extraction through cross-document inference. / Ji, Heng; Grishman, Ralph.

ACL-08: HLT - 46th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference. 2008. p. 254-262.

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

Ji, H & Grishman, R 2008, Refining event extraction through cross-document inference. in ACL-08: HLT - 46th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference. pp. 254-262, 46th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, ACL-08: HLT, Columbus, OH, United States, 6/15/08.
Ji H, Grishman R. Refining event extraction through cross-document inference. In ACL-08: HLT - 46th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference. 2008. p. 254-262
Ji, Heng ; Grishman, Ralph. / Refining event extraction through cross-document inference. ACL-08: HLT - 46th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference. 2008. pp. 254-262
@inproceedings{775a3a688d0a4736ad7954edb548eaa2,
title = "Refining event extraction through cross-document inference",
abstract = "We apply the hypothesis of {"}One Sense Per Discourse{"} (Yarowsky, 1995) to information extraction (IE), and extend the scope of {"}discourse{"} from one single document to a cluster of topically-related documents. We employ a similar approach to propagate consistent event arguments across sentences and documents. Combining global evidence from related documents with local decisions, we design a simple scheme to conduct cross-document inference for improving the ACE event extraction task1. Without using any additional labeled data this new approach obtained 7.6{\%} higher F-Measure in trigger labeling and 6{\%} higher F-Measure in argument labeling over a state-of-the-art IE system which extracts events independently for each sentence.",
author = "Heng Ji and Ralph Grishman",
year = "2008",
language = "English (US)",
isbn = "9781932432046",
pages = "254--262",
booktitle = "ACL-08: HLT - 46th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference",

}

TY - GEN

T1 - Refining event extraction through cross-document inference

AU - Ji, Heng

AU - Grishman, Ralph

PY - 2008

Y1 - 2008

N2 - We apply the hypothesis of "One Sense Per Discourse" (Yarowsky, 1995) to information extraction (IE), and extend the scope of "discourse" from one single document to a cluster of topically-related documents. We employ a similar approach to propagate consistent event arguments across sentences and documents. Combining global evidence from related documents with local decisions, we design a simple scheme to conduct cross-document inference for improving the ACE event extraction task1. Without using any additional labeled data this new approach obtained 7.6% higher F-Measure in trigger labeling and 6% higher F-Measure in argument labeling over a state-of-the-art IE system which extracts events independently for each sentence.

AB - We apply the hypothesis of "One Sense Per Discourse" (Yarowsky, 1995) to information extraction (IE), and extend the scope of "discourse" from one single document to a cluster of topically-related documents. We employ a similar approach to propagate consistent event arguments across sentences and documents. Combining global evidence from related documents with local decisions, we design a simple scheme to conduct cross-document inference for improving the ACE event extraction task1. Without using any additional labeled data this new approach obtained 7.6% higher F-Measure in trigger labeling and 6% higher F-Measure in argument labeling over a state-of-the-art IE system which extracts events independently for each sentence.

UR - http://www.scopus.com/inward/record.url?scp=84859903043&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84859903043&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84859903043

SN - 9781932432046

SP - 254

EP - 262

BT - ACL-08: HLT - 46th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference

ER -