Improving word segmentation by simultaneously learning phonotactics

Daniel Blanchard; Jeffrey Heinz

doi:10.3115/1596324.1596336

Improving word segmentation by simultaneously learning phonotactics

Daniel Blanchard
, Jeffrey Heinz

Linguistics

University of Delaware

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

9 Scopus citations

Abstract

The most accurate unsupervised word segmentation systems that are currently available (Brent, 1999; Venkataraman, 2001; Goldwater, 2007) use a simple unigram model of phonotactics. While this simplifies some of the calculations, it overlooks cues that infant language acquisition researchers have shown to be useful for segmentation (Mattys et al., 1999; Mattys and Jusczyk, 2001). Here we explore the utility of using bigram and trigram phonotactic models by enhancing Brent's (1999) MBDP-1 algorithm. The results show the improved MBDP-Phon model outperforms other unsupervised word segmentation systems (e.g., Brent, 1999; Venkataraman, 2001; Goldwater, 2007).

Original language	English
Title of host publication	CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning
Publisher	Association for Computational Linguistics (ACL)
Pages	65-72
Number of pages	8
ISBN (Print)	1905593481, 9781905593484
DOIs	https://doi.org/10.3115/1596324.1596336
State	Published - 2008
Event	12th Conference on Computational Natural Language Learning, CoNLL 2008 - Manchester, United Kingdom Duration: Aug 16 2008 → Aug 17 2008

Publication series

Name	CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning

Conference

Conference	12th Conference on Computational Natural Language Learning, CoNLL 2008
Country/Territory	United Kingdom
City	Manchester
Period	08/16/08 → 08/17/08

Access to Document

10.3115/1596324.1596336

Cite this

Blanchard, D., & Heinz, J. (2008). Improving word segmentation by simultaneously learning phonotactics. In CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning (pp. 65-72). (CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning). Association for Computational Linguistics (ACL). https://doi.org/10.3115/1596324.1596336

@inproceedings{51bc165107814b70971b41b11e6d2741,

title = "Improving word segmentation by simultaneously learning phonotactics",

abstract = "The most accurate unsupervised word segmentation systems that are currently available (Brent, 1999; Venkataraman, 2001; Goldwater, 2007) use a simple unigram model of phonotactics. While this simplifies some of the calculations, it overlooks cues that infant language acquisition researchers have shown to be useful for segmentation (Mattys et al., 1999; Mattys and Jusczyk, 2001). Here we explore the utility of using bigram and trigram phonotactic models by enhancing Brent's (1999) MBDP-1 algorithm. The results show the improved MBDP-Phon model outperforms other unsupervised word segmentation systems (e.g., Brent, 1999; Venkataraman, 2001; Goldwater, 2007).",

author = "Daniel Blanchard and Jeffrey Heinz",

year = "2008",

doi = "10.3115/1596324.1596336",

language = "English",

isbn = "1905593481",

series = "CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning",

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

pages = "65--72",

booktitle = "CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning",

note = "12th Conference on Computational Natural Language Learning, CoNLL 2008 ; Conference date: 16-08-2008 Through 17-08-2008",

}

Blanchard, D & Heinz, J 2008, Improving word segmentation by simultaneously learning phonotactics. in CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning. CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning, Association for Computational Linguistics (ACL), pp. 65-72, 12th Conference on Computational Natural Language Learning, CoNLL 2008, Manchester, United Kingdom, 08/16/08. https://doi.org/10.3115/1596324.1596336

Improving word segmentation by simultaneously learning phonotactics. / Blanchard, Daniel; Heinz, Jeffrey.
CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning. Association for Computational Linguistics (ACL), 2008. p. 65-72 (CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning).

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

TY - GEN

T1 - Improving word segmentation by simultaneously learning phonotactics

AU - Blanchard, Daniel

AU - Heinz, Jeffrey

PY - 2008

Y1 - 2008

N2 - The most accurate unsupervised word segmentation systems that are currently available (Brent, 1999; Venkataraman, 2001; Goldwater, 2007) use a simple unigram model of phonotactics. While this simplifies some of the calculations, it overlooks cues that infant language acquisition researchers have shown to be useful for segmentation (Mattys et al., 1999; Mattys and Jusczyk, 2001). Here we explore the utility of using bigram and trigram phonotactic models by enhancing Brent's (1999) MBDP-1 algorithm. The results show the improved MBDP-Phon model outperforms other unsupervised word segmentation systems (e.g., Brent, 1999; Venkataraman, 2001; Goldwater, 2007).

AB - The most accurate unsupervised word segmentation systems that are currently available (Brent, 1999; Venkataraman, 2001; Goldwater, 2007) use a simple unigram model of phonotactics. While this simplifies some of the calculations, it overlooks cues that infant language acquisition researchers have shown to be useful for segmentation (Mattys et al., 1999; Mattys and Jusczyk, 2001). Here we explore the utility of using bigram and trigram phonotactic models by enhancing Brent's (1999) MBDP-1 algorithm. The results show the improved MBDP-Phon model outperforms other unsupervised word segmentation systems (e.g., Brent, 1999; Venkataraman, 2001; Goldwater, 2007).

UR - https://www.scopus.com/pages/publications/77952548445

U2 - 10.3115/1596324.1596336

DO - 10.3115/1596324.1596336

M3 - Conference contribution

AN - SCOPUS:77952548445

SN - 1905593481

SN - 9781905593484

T3 - CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning

SP - 65

EP - 72

BT - CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning

PB - Association for Computational Linguistics (ACL)

T2 - 12th Conference on Computational Natural Language Learning, CoNLL 2008

Y2 - 16 August 2008 through 17 August 2008

ER -

Blanchard D, Heinz J. Improving word segmentation by simultaneously learning phonotactics. In CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning. Association for Computational Linguistics (ACL). 2008. p. 65-72. (CoNLL 2008 - Proceedings of the Twelfth Conference on Computational Natural Language Learning). doi: 10.3115/1596324.1596336

Improving word segmentation by simultaneously learning phonotactics

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this