Adapting coreference resolution for narrative processing

Quynh Ngoc Thi Do; Steven Bethard; Marie Francine Moens

doi:10.18653/v1/d15-1271

Adapting coreference resolution for narrative processing

Quynh Ngoc Thi Do, Steven Bethard, Marie Francine Moens

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

5 Scopus citations

Abstract

Domain adaptation is a challenge for supervised NLP systems because of expensive and time-consuming manual annotated resources. We present a novel method to adapt a supervised coreference resolution system trained on newswire to short narrative stories without retraining the system. The idea is to perform inference via an Integer Linear Programming (ILP) formulation with the features of narratives adopted as soft constraints. When testing on the UMIREC¹ and N2² corpora with the-stateof-the-art Berkeley coreference resolution system trained on OntoNotes³, our inference substantially outperforms the original inference on the CoNLL 2011 metric.

Original language	English (US)
Title of host publication	Conference Proceedings - EMNLP 2015
Subtitle of host publication	Conference on Empirical Methods in Natural Language Processing
Publisher	Association for Computational Linguistics (ACL)
Pages	2262-2267
Number of pages	6
ISBN (Electronic)	9781941643327
DOIs	https://doi.org/10.18653/v1/d15-1271
State	Published - 2015
Externally published	Yes
Event	Conference on Empirical Methods in Natural Language Processing, EMNLP 2015 - Lisbon, Portugal Duration: Sep 17 2015 → Sep 21 2015

Publication series

Name	Conference Proceedings - EMNLP 2015: Conference on Empirical Methods in Natural Language Processing

Conference

Conference	Conference on Empirical Methods in Natural Language Processing, EMNLP 2015
Country/Territory	Portugal
City	Lisbon
Period	9/17/15 → 9/21/15

ASJC Scopus subject areas

Computational Theory and Mathematics
Computer Science Applications
Information Systems

Access to Document

10.18653/v1/d15-1271

Cite this

Do, Q. N. T., Bethard, S., & Moens, M. F. (2015). Adapting coreference resolution for narrative processing. In Conference Proceedings - EMNLP 2015: Conference on Empirical Methods in Natural Language Processing (pp. 2262-2267). (Conference Proceedings - EMNLP 2015: Conference on Empirical Methods in Natural Language Processing). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/d15-1271

Adapting coreference resolution for narrative processing. / Do, Quynh Ngoc Thi; Bethard, Steven; Moens, Marie Francine.
Conference Proceedings - EMNLP 2015: Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics (ACL), 2015. p. 2262-2267 (Conference Proceedings - EMNLP 2015: Conference on Empirical Methods in Natural Language Processing).

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

Do, QNT, Bethard, S & Moens, MF 2015, Adapting coreference resolution for narrative processing. in Conference Proceedings - EMNLP 2015: Conference on Empirical Methods in Natural Language Processing. Conference Proceedings - EMNLP 2015: Conference on Empirical Methods in Natural Language Processing, Association for Computational Linguistics (ACL), pp. 2262-2267, Conference on Empirical Methods in Natural Language Processing, EMNLP 2015, Lisbon, Portugal, 9/17/15. https://doi.org/10.18653/v1/d15-1271

Do QNT, Bethard S, Moens MF. Adapting coreference resolution for narrative processing. In Conference Proceedings - EMNLP 2015: Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics (ACL). 2015. p. 2262-2267. (Conference Proceedings - EMNLP 2015: Conference on Empirical Methods in Natural Language Processing). doi: 10.18653/v1/d15-1271

Do, Quynh Ngoc Thi ; Bethard, Steven ; Moens, Marie Francine. / Adapting coreference resolution for narrative processing. Conference Proceedings - EMNLP 2015: Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics (ACL), 2015. pp. 2262-2267 (Conference Proceedings - EMNLP 2015: Conference on Empirical Methods in Natural Language Processing).

@inproceedings{5969b23c25ec4d79865ffa69bc710a8f,

title = "Adapting coreference resolution for narrative processing",

abstract = "Domain adaptation is a challenge for supervised NLP systems because of expensive and time-consuming manual annotated resources. We present a novel method to adapt a supervised coreference resolution system trained on newswire to short narrative stories without retraining the system. The idea is to perform inference via an Integer Linear Programming (ILP) formulation with the features of narratives adopted as soft constraints. When testing on the UMIREC1 and N22 corpora with the-stateof-the-art Berkeley coreference resolution system trained on OntoNotes3, our inference substantially outperforms the original inference on the CoNLL 2011 metric.",

author = "Do, {Quynh Ngoc Thi} and Steven Bethard and Moens, {Marie Francine}",

note = "Publisher Copyright: {\textcopyright} 2015 Association for Computational Linguistics.; Conference on Empirical Methods in Natural Language Processing, EMNLP 2015 ; Conference date: 17-09-2015 Through 21-09-2015",

year = "2015",

doi = "10.18653/v1/d15-1271",

language = "English (US)",

series = "Conference Proceedings - EMNLP 2015: Conference on Empirical Methods in Natural Language Processing",

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

pages = "2262--2267",

booktitle = "Conference Proceedings - EMNLP 2015",

}

TY - GEN

T1 - Adapting coreference resolution for narrative processing

AU - Do, Quynh Ngoc Thi

AU - Bethard, Steven

AU - Moens, Marie Francine

PY - 2015

Y1 - 2015

N2 - Domain adaptation is a challenge for supervised NLP systems because of expensive and time-consuming manual annotated resources. We present a novel method to adapt a supervised coreference resolution system trained on newswire to short narrative stories without retraining the system. The idea is to perform inference via an Integer Linear Programming (ILP) formulation with the features of narratives adopted as soft constraints. When testing on the UMIREC1 and N22 corpora with the-stateof-the-art Berkeley coreference resolution system trained on OntoNotes3, our inference substantially outperforms the original inference on the CoNLL 2011 metric.

AB - Domain adaptation is a challenge for supervised NLP systems because of expensive and time-consuming manual annotated resources. We present a novel method to adapt a supervised coreference resolution system trained on newswire to short narrative stories without retraining the system. The idea is to perform inference via an Integer Linear Programming (ILP) formulation with the features of narratives adopted as soft constraints. When testing on the UMIREC1 and N22 corpora with the-stateof-the-art Berkeley coreference resolution system trained on OntoNotes3, our inference substantially outperforms the original inference on the CoNLL 2011 metric.

UR - http://www.scopus.com/inward/record.url?scp=84959887520&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84959887520&partnerID=8YFLogxK

U2 - 10.18653/v1/d15-1271

DO - 10.18653/v1/d15-1271

M3 - Conference contribution

AN - SCOPUS:84959887520

T3 - Conference Proceedings - EMNLP 2015: Conference on Empirical Methods in Natural Language Processing

SP - 2262

EP - 2267

BT - Conference Proceedings - EMNLP 2015

PB - Association for Computational Linguistics (ACL)

T2 - Conference on Empirical Methods in Natural Language Processing, EMNLP 2015

Y2 - 17 September 2015 through 21 September 2015

ER -

Adapting coreference resolution for narrative processing

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this