Character profiling in low-resource language documents

Tak-sum Wong; John Lee

doi:10.1145/3372124.3372129

Character profiling in low-resource language documents

Department of Linguistics and Translation

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

Abstract

This paper focuses on automatic character profiling — connecting “who”, “what” and “when” — in literary documents. This task is especially challenging for low-resource languages, since off-the-shelf tools for named entity recognition, syntactic parsing and other natural language processing tasks are rarely available. We investigate the impact of human annotation on automatic profiling. Based on a Medieval Chinese corpus, experimental results show that even a relatively small amount of word segmentation, part-of-speech and dependency annotation can improve accuracy in named entity recognition and in identifying character-verb associations, but not character-toponym associations.

Original language	English
Title of host publication	ADCS '19: Proceedings of the 24th Australasian Document Computing Symposium
Editors	Gianluca Demartini, Paul Thomas
Publisher	Association for Computing Machinery
ISBN (Electronic)	9781450377669
DOIs	https://doi.org/10.1145/3372124.3372129
Publication status	Published - Dec 2019
Event	24^th Australasian Document Computing Symposium (ADCS 2019) - University of Technology, Sydney, Australia Duration: 5 Dec 2019 → 6 Dec 2019 http://adcs-conference.org/2019/

Publication series

Name	ACM International Conference Proceeding Series

Conference

Conference	24^th Australasian Document Computing Symposium (ADCS 2019)
Abbreviated title	ADCS 2019
Place	Australia
City	Sydney
Period	5/12/19 → 6/12/19
Internet address	http://adcs-conference.org/2019/

Bibliographical note

Full text of this publication does not contain sufficient affiliation information. With consent from the author(s) concerned, the Research Unit(s) information for this record is based on the existing academic department affiliation of the author(s).

Research Keywords

Dependency parsing
Information extraction
Low-resource language
Medieval Chinese
Named entity recognition

Access Output

10.1145/3372124.3372129

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{d13752309d2642879b0152cd948352ae,

title = "Character profiling in low-resource language documents",

abstract = "This paper focuses on automatic character profiling — connecting “who”, “what” and “when” — in literary documents. This task is especially challenging for low-resource languages, since off-the-shelf tools for named entity recognition, syntactic parsing and other natural language processing tasks are rarely available. We investigate the impact of human annotation on automatic profiling. Based on a Medieval Chinese corpus, experimental results show that even a relatively small amount of word segmentation, part-of-speech and dependency annotation can improve accuracy in named entity recognition and in identifying character-verb associations, but not character-toponym associations.",

keywords = "Dependency parsing, Information extraction, Low-resource language, Medieval Chinese, Named entity recognition",

author = "Tak-sum Wong and John Lee",

note = "Full text of this publication does not contain sufficient affiliation information. With consent from the author(s) concerned, the Research Unit(s) information for this record is based on the existing academic department affiliation of the author(s).; 24<sup>th</sup> Australasian Document Computing Symposium (ADCS 2019), ADCS 2019 ; Conference date: 05-12-2019 Through 06-12-2019",

year = "2019",

month = dec,

doi = "10.1145/3372124.3372129",

language = "English",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

editor = "Gianluca Demartini and Paul Thomas",

booktitle = "ADCS '19: Proceedings of the 24th Australasian Document Computing Symposium",

address = "United States",

url = "http://adcs-conference.org/2019/",

}

Wong, T & Lee, J 2019, Character profiling in low-resource language documents. in G Demartini & P Thomas (eds), ADCS '19: Proceedings of the 24th Australasian Document Computing Symposium., 5, ACM International Conference Proceeding Series, Association for Computing Machinery, 24^th Australasian Document Computing Symposium (ADCS 2019), Sydney, Australia, 5/12/19. https://doi.org/10.1145/3372124.3372129

Character profiling in low-resource language documents. / Wong, Tak-sum ; Lee, John.
ADCS '19: Proceedings of the 24th Australasian Document Computing Symposium. ed. / Gianluca Demartini; Paul Thomas. Association for Computing Machinery, 2019. 5 (ACM International Conference Proceeding Series).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Character profiling in low-resource language documents

AU - Wong, Tak-sum

AU - Lee, John

N1 - Full text of this publication does not contain sufficient affiliation information. With consent from the author(s) concerned, the Research Unit(s) information for this record is based on the existing academic department affiliation of the author(s).

PY - 2019/12

Y1 - 2019/12

N2 - This paper focuses on automatic character profiling — connecting “who”, “what” and “when” — in literary documents. This task is especially challenging for low-resource languages, since off-the-shelf tools for named entity recognition, syntactic parsing and other natural language processing tasks are rarely available. We investigate the impact of human annotation on automatic profiling. Based on a Medieval Chinese corpus, experimental results show that even a relatively small amount of word segmentation, part-of-speech and dependency annotation can improve accuracy in named entity recognition and in identifying character-verb associations, but not character-toponym associations.

AB - This paper focuses on automatic character profiling — connecting “who”, “what” and “when” — in literary documents. This task is especially challenging for low-resource languages, since off-the-shelf tools for named entity recognition, syntactic parsing and other natural language processing tasks are rarely available. We investigate the impact of human annotation on automatic profiling. Based on a Medieval Chinese corpus, experimental results show that even a relatively small amount of word segmentation, part-of-speech and dependency annotation can improve accuracy in named entity recognition and in identifying character-verb associations, but not character-toponym associations.

KW - Dependency parsing

KW - Information extraction

KW - Low-resource language

KW - Medieval Chinese

KW - Named entity recognition

UR - http://www.scopus.com/inward/record.url?scp=85077712610&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85077712610&origin=recordpage

U2 - 10.1145/3372124.3372129

DO - 10.1145/3372124.3372129

M3 - RGC 32 - Refereed conference paper (with host publication)

T3 - ACM International Conference Proceeding Series

BT - ADCS '19: Proceedings of the 24th Australasian Document Computing Symposium

A2 - Demartini, Gianluca

A2 - Thomas, Paul

PB - Association for Computing Machinery

T2 - 24<sup>th</sup> Australasian Document Computing Symposium (ADCS 2019)

Y2 - 5 December 2019 through 6 December 2019

ER -

Character profiling in low-resource language documents

Abstract

Publication series

Conference

Bibliographical note

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this