Automatic word segmentation for spoken Cantonese

Suk Yee Roxana Fung; Brigitte Bigi

doi:10.1109/ICSDA.2015.7357891

Automatic word segmentation for spoken Cantonese

Suk Yee Roxana Fung, Brigitte Bigi

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

9 Citations (Scopus)

Abstract

Though Cantonese is the most influential variety of Chinese other than Mandarin, there are only a limited number of Cantonese corpora available for linguistic studies. Among the essential steps of building a corpus, word segmentation is a necessary but highly challenging task due to the lack of clear word boundary in Cantonese. This paper reports the construction and evaluation of an open-source automatic Cantonese word segmenter developed for Cantonese. The tool is a component of the multilingual SPPAS program designed to be used directly by linguists. It is a free software distributed under a GPL license. The effectiveness of the tool was evaluated by comparing the result of segmenting some samples of a spoken Cantonese corpus manually and automatically using the tool developed. High precision and recall were found in our study. Upon completion, the tool would definitely promote the development of more Cantonese corpora for language related studies.

Original language	English
Title of host publication	2015 International Conference Oriental COCOSDA - Held jointly with 2015 Conference on Asian Spoken Language Research and Evaluation, O-COCOSDA/CASLRE 2015 - Proceedings
Publisher	IEEE
Pages	196-201
Number of pages	6
ISBN (Electronic)	9781467382793
DOIs	https://doi.org/10.1109/ICSDA.2015.7357891
Publication status	Published - 14 Dec 2015
Event	18th Annual International Conference Oriental COCOSDA, O-COCOSDA 2015 - Held jointly with 2015 Conference on Asian Spoken Language Research and Evaluation, CASLRE 2015 - Shanghai Jiao Tong University, Shanghai, China Duration: 28 Oct 2015 → 30 Oct 2015

Conference

Conference	18th Annual International Conference Oriental COCOSDA, O-COCOSDA 2015 - Held jointly with 2015 Conference on Asian Spoken Language Research and Evaluation, CASLRE 2015
Country/Territory	China
City	Shanghai
Period	28/10/15 → 30/10/15

Keywords

automatic
Cantonese
corpus
segmentation
software

ASJC Scopus subject areas

Linguistics and Language
Computer Science Applications
Software
Language and Linguistics

More information

10.1109/ICSDA.2015.7357891

Cite this

@inproceedings{c4c039097b804729865de1d2b3191488,

title = "Automatic word segmentation for spoken Cantonese",

abstract = "Though Cantonese is the most influential variety of Chinese other than Mandarin, there are only a limited number of Cantonese corpora available for linguistic studies. Among the essential steps of building a corpus, word segmentation is a necessary but highly challenging task due to the lack of clear word boundary in Cantonese. This paper reports the construction and evaluation of an open-source automatic Cantonese word segmenter developed for Cantonese. The tool is a component of the multilingual SPPAS program designed to be used directly by linguists. It is a free software distributed under a GPL license. The effectiveness of the tool was evaluated by comparing the result of segmenting some samples of a spoken Cantonese corpus manually and automatically using the tool developed. High precision and recall were found in our study. Upon completion, the tool would definitely promote the development of more Cantonese corpora for language related studies.",

keywords = "automatic, Cantonese, corpus, segmentation, software",

author = "Fung, {Suk Yee Roxana} and Brigitte Bigi",

year = "2015",

month = dec,

day = "14",

doi = "10.1109/ICSDA.2015.7357891",

language = "English",

pages = "196--201",

booktitle = "2015 International Conference Oriental COCOSDA - Held jointly with 2015 Conference on Asian Spoken Language Research and Evaluation, O-COCOSDA/CASLRE 2015 - Proceedings",

publisher = "IEEE",

address = "United States",

note = "18th Annual International Conference Oriental COCOSDA, O-COCOSDA 2015 - Held jointly with 2015 Conference on Asian Spoken Language Research and Evaluation, CASLRE 2015 ; Conference date: 28-10-2015 Through 30-10-2015",

}

Fung, SYR & Bigi, B 2015, Automatic word segmentation for spoken Cantonese. in 2015 International Conference Oriental COCOSDA - Held jointly with 2015 Conference on Asian Spoken Language Research and Evaluation, O-COCOSDA/CASLRE 2015 - Proceedings., 7357891, IEEE, pp. 196-201, 18th Annual International Conference Oriental COCOSDA, O-COCOSDA 2015 - Held jointly with 2015 Conference on Asian Spoken Language Research and Evaluation, CASLRE 2015, Shanghai, China, 28/10/15. https://doi.org/10.1109/ICSDA.2015.7357891

Automatic word segmentation for spoken Cantonese. / Fung, Suk Yee Roxana; Bigi, Brigitte.
2015 International Conference Oriental COCOSDA - Held jointly with 2015 Conference on Asian Spoken Language Research and Evaluation, O-COCOSDA/CASLRE 2015 - Proceedings. IEEE, 2015. p. 196-201 7357891.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Automatic word segmentation for spoken Cantonese

AU - Fung, Suk Yee Roxana

AU - Bigi, Brigitte

PY - 2015/12/14

Y1 - 2015/12/14

N2 - Though Cantonese is the most influential variety of Chinese other than Mandarin, there are only a limited number of Cantonese corpora available for linguistic studies. Among the essential steps of building a corpus, word segmentation is a necessary but highly challenging task due to the lack of clear word boundary in Cantonese. This paper reports the construction and evaluation of an open-source automatic Cantonese word segmenter developed for Cantonese. The tool is a component of the multilingual SPPAS program designed to be used directly by linguists. It is a free software distributed under a GPL license. The effectiveness of the tool was evaluated by comparing the result of segmenting some samples of a spoken Cantonese corpus manually and automatically using the tool developed. High precision and recall were found in our study. Upon completion, the tool would definitely promote the development of more Cantonese corpora for language related studies.

AB - Though Cantonese is the most influential variety of Chinese other than Mandarin, there are only a limited number of Cantonese corpora available for linguistic studies. Among the essential steps of building a corpus, word segmentation is a necessary but highly challenging task due to the lack of clear word boundary in Cantonese. This paper reports the construction and evaluation of an open-source automatic Cantonese word segmenter developed for Cantonese. The tool is a component of the multilingual SPPAS program designed to be used directly by linguists. It is a free software distributed under a GPL license. The effectiveness of the tool was evaluated by comparing the result of segmenting some samples of a spoken Cantonese corpus manually and automatically using the tool developed. High precision and recall were found in our study. Upon completion, the tool would definitely promote the development of more Cantonese corpora for language related studies.

KW - automatic

KW - Cantonese

KW - corpus

KW - segmentation

KW - software

UR - http://www.scopus.com/inward/record.url?scp=84962900923&partnerID=8YFLogxK

U2 - 10.1109/ICSDA.2015.7357891

DO - 10.1109/ICSDA.2015.7357891

M3 - Conference article published in proceeding or book

SP - 196

EP - 201

BT - 2015 International Conference Oriental COCOSDA - Held jointly with 2015 Conference on Asian Spoken Language Research and Evaluation, O-COCOSDA/CASLRE 2015 - Proceedings

PB - IEEE

T2 - 18th Annual International Conference Oriental COCOSDA, O-COCOSDA 2015 - Held jointly with 2015 Conference on Asian Spoken Language Research and Evaluation, CASLRE 2015

Y2 - 28 October 2015 through 30 October 2015

ER -

Automatic word segmentation for spoken Cantonese

Abstract

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this