Cross-document coreference: An approach to capturing coreference without context

Kristin Wright-Bettner; Martha Palmer; Guergana Savova; Piet de Groen; Timothy Miller

Cross-document coreference: An approach to capturing coreference without context

Kristin Wright-Bettner
, Martha Palmer
, Guergana Savova
, Piet de Groen
, Timothy Miller

Medicine - Gastro, Hepatology, Nutrition Division

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

9 Scopus citations

Abstract

In this paper, we discuss a cross-document coreference annotation schema that we developed to further automatic extraction of timelines in the clinical domain. Lexical senses and coreference choices are determined largely by context, but cross-document work requires reasoning across contexts that are not necessarily coherent. We found that an annotation approach that relies less on context-guided annotator intuitions and more on schematic rules was most effective in creating meaningful and consistent cross-document relations.

Original language	English (US)
Title of host publication	LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings
Publisher	Association for Computational Linguistics (ACL)
Pages	1-10
Number of pages	10
ISBN (Electronic)	9781950737772
State	Published - 2019
Event	10th International Workshop on Health Text Mining and Information Analysis, LOUHI@EMNLP 2019 - Hong Kong, China Duration: Nov 3 2019 → …

Publication series

Name	LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings

Conference

Conference	10th International Workshop on Health Text Mining and Information Analysis, LOUHI@EMNLP 2019
Country/Territory	China
City	Hong Kong
Period	11/3/19 → …

Bibliographical note

Funding Information:
The work was supported by funding R01LM010090 from the National Library Of Medicine. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Library Of Medicine or the National Institute of Health.

Funding Information:
The work was supported by funding R01LM010090 from the National Library Of Medicine. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Library Of Medicine or the National Institute of Health. We would like to thank: Dana Green especially for annotation and insightful annotation input; Ahmed Elsayed and Dave Harris for medical annotation and advice; James Martin for schema development advice; Wei-Te Chen and Skatje Myers for technical support; Michael Regan, Matthew Oh, Hayley Coniglio, Samuel Beer, and Jameson Ducey for annotation; and Adam Wiemerslage for IAA and post-processing scripts.

Publisher Copyright:
© 2019 Association for Computational Linguistics

Find It

Find It at U of M Libraries

Cite this

Wright-Bettner, K., Palmer, M., Savova, G., de Groen, P., & Miller, T. (2019). Cross-document coreference: An approach to capturing coreference without context. In LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings (pp. 1-10). (LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings). Association for Computational Linguistics (ACL).

Cross-document coreference: An approach to capturing coreference without context. / Wright-Bettner, Kristin; Palmer, Martha; Savova, Guergana et al.
LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings. Association for Computational Linguistics (ACL), 2019. p. 1-10 (LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings).

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

Wright-Bettner, K, Palmer, M, Savova, G, de Groen, P & Miller, T 2019, Cross-document coreference: An approach to capturing coreference without context. in LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings. LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings, Association for Computational Linguistics (ACL), pp. 1-10, 10th International Workshop on Health Text Mining and Information Analysis, LOUHI@EMNLP 2019, Hong Kong, China, 11/3/19.

Wright-Bettner K, Palmer M, Savova G, de Groen P, Miller T. Cross-document coreference: An approach to capturing coreference without context. In LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings. Association for Computational Linguistics (ACL). 2019. p. 1-10. (LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings).

Wright-Bettner, Kristin ; Palmer, Martha ; Savova, Guergana et al. / Cross-document coreference : An approach to capturing coreference without context. LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings. Association for Computational Linguistics (ACL), 2019. pp. 1-10 (LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings).

@inproceedings{11a79507554444d4ad14d0325fbdec0c,

title = "Cross-document coreference: An approach to capturing coreference without context",

abstract = "In this paper, we discuss a cross-document coreference annotation schema that we developed to further automatic extraction of timelines in the clinical domain. Lexical senses and coreference choices are determined largely by context, but cross-document work requires reasoning across contexts that are not necessarily coherent. We found that an annotation approach that relies less on context-guided annotator intuitions and more on schematic rules was most effective in creating meaningful and consistent cross-document relations.",

author = "Kristin Wright-Bettner and Martha Palmer and Guergana Savova and \{de Groen\}, Piet and Timothy Miller",

note = "Funding Information: The work was supported by funding R01LM010090 from the National Library Of Medicine. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Library Of Medicine or the National Institute of Health. Funding Information: The work was supported by funding R01LM010090 from the National Library Of Medicine. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Library Of Medicine or the National Institute of Health. We would like to thank: Dana Green especially for annotation and insightful annotation input; Ahmed Elsayed and Dave Harris for medical annotation and advice; James Martin for schema development advice; Wei-Te Chen and Skatje Myers for technical support; Michael Regan, Matthew Oh, Hayley Coniglio, Samuel Beer, and Jameson Ducey for annotation; and Adam Wiemerslage for IAA and post-processing scripts. Publisher Copyright: {\textcopyright} 2019 Association for Computational Linguistics; 10th International Workshop on Health Text Mining and Information Analysis, LOUHI@EMNLP 2019 ; Conference date: 03-11-2019",

year = "2019",

language = "English (US)",

series = "LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings",

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

pages = "1--10",

booktitle = "LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings",

}

TY - GEN

T1 - Cross-document coreference

T2 - 10th International Workshop on Health Text Mining and Information Analysis, LOUHI@EMNLP 2019

AU - Wright-Bettner, Kristin

AU - Palmer, Martha

AU - Savova, Guergana

AU - de Groen, Piet

AU - Miller, Timothy

N1 - Funding Information: The work was supported by funding R01LM010090 from the National Library Of Medicine. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Library Of Medicine or the National Institute of Health. Funding Information: The work was supported by funding R01LM010090 from the National Library Of Medicine. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Library Of Medicine or the National Institute of Health. We would like to thank: Dana Green especially for annotation and insightful annotation input; Ahmed Elsayed and Dave Harris for medical annotation and advice; James Martin for schema development advice; Wei-Te Chen and Skatje Myers for technical support; Michael Regan, Matthew Oh, Hayley Coniglio, Samuel Beer, and Jameson Ducey for annotation; and Adam Wiemerslage for IAA and post-processing scripts. Publisher Copyright: © 2019 Association for Computational Linguistics

PY - 2019

Y1 - 2019

N2 - In this paper, we discuss a cross-document coreference annotation schema that we developed to further automatic extraction of timelines in the clinical domain. Lexical senses and coreference choices are determined largely by context, but cross-document work requires reasoning across contexts that are not necessarily coherent. We found that an annotation approach that relies less on context-guided annotator intuitions and more on schematic rules was most effective in creating meaningful and consistent cross-document relations.

AB - In this paper, we discuss a cross-document coreference annotation schema that we developed to further automatic extraction of timelines in the clinical domain. Lexical senses and coreference choices are determined largely by context, but cross-document work requires reasoning across contexts that are not necessarily coherent. We found that an annotation approach that relies less on context-guided annotator intuitions and more on schematic rules was most effective in creating meaningful and consistent cross-document relations.

UR - https://www.scopus.com/pages/publications/85115853510

UR - https://www.scopus.com/pages/publications/85115853510#tab=citedBy

M3 - Conference contribution

AN - SCOPUS:85115853510

T3 - LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings

SP - 1

EP - 10

BT - LOUHI@EMNLP 2019 - 10th International Workshop on Health Text Mining and Information Analysis, Proceedings

PB - Association for Computational Linguistics (ACL)

Y2 - 3 November 2019

ER -

Cross-document coreference: An approach to capturing coreference without context

Abstract

Publication series

Conference

Bibliographical note

Find It

Other files and links

Fingerprint

Cite this