Unsupervised text segmentation based on native language characteristics

Shervin Malmasi; Mark Dras; Mark Johnson; Lan Du; Magdalena Wolska

doi:10.18653/v1/P17-1134

Unsupervised text segmentation based on native language characteristics

Shervin Malmasi, Mark Dras, Mark Johnson, Lan Du, Magdalena Wolska

School of Computing

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

8 Citations (Scopus)

36 Downloads (Pure)

Abstract

Most work on segmenting text does so on the basis of topic changes, but it can be of interest to segment by other, stylistically expressed characteristics such as change of authorship or native language. We propose a Bayesian unsupervised text segmentation approach to the latter. While baseline models achieve essentially random segmentation on our task, indicating its difficulty, a Bayesian model that incorporates appropriately compact language models and alternating asymmetric priors can achieve scores on the standard metrics around halfway to perfect segmentation.

Original language	English
Title of host publication	ACL 2017
Subtitle of host publication	Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Long Papers)
Editors	Regina Barzilay, Min-Yen Kan
Place of Publication	Stroudsburg, PA
Publisher	Association for Computational Linguistics (ACL)
Pages	1457-1469
Number of pages	13
Volume	1
ISBN (Electronic)	9781945626753
DOIs	https://doi.org/10.18653/v1/P17-1134
Publication status	Published - 2017
Event	55th Annual Meeting of the Association for Computational Linguistics, ACL 2017 - Vancouver, Canada Duration: 30 Jul 2017 → 4 Aug 2017

Conference

Conference	55th Annual Meeting of the Association for Computational Linguistics, ACL 2017
Country/Territory	Canada
City	Vancouver
Period	30/07/17 → 4/08/17

Bibliographical note

Copyright the Publisher. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

Access to Document

10.18653/v1/P17-1134

Publisher version (open access)Final published version, 340 KBLicence: CC BY

Cite this

Malmasi, S., Dras, M., Johnson, M., Du, L., & Wolska, M. (2017). Unsupervised text segmentation based on native language characteristics. In R. Barzilay, & M.-Y. Kan (Eds.), ACL 2017: Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Long Papers) (Vol. 1, pp. 1457-1469). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/P17-1134

Malmasi, Shervin ; Dras, Mark ; Johnson, Mark et al. / Unsupervised text segmentation based on native language characteristics. ACL 2017: Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Long Papers). editor / Regina Barzilay ; Min-Yen Kan. Vol. 1 Stroudsburg, PA : Association for Computational Linguistics (ACL), 2017. pp. 1457-1469

@inproceedings{9b50a1dfa7ce4f09ae98f3623a65c1c2,

title = "Unsupervised text segmentation based on native language characteristics",

abstract = "Most work on segmenting text does so on the basis of topic changes, but it can be of interest to segment by other, stylistically expressed characteristics such as change of authorship or native language. We propose a Bayesian unsupervised text segmentation approach to the latter. While baseline models achieve essentially random segmentation on our task, indicating its difficulty, a Bayesian model that incorporates appropriately compact language models and alternating asymmetric priors can achieve scores on the standard metrics around halfway to perfect segmentation.",

author = "Shervin Malmasi and Mark Dras and Mark Johnson and Lan Du and Magdalena Wolska",

note = "Copyright the Publisher. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.; 55th Annual Meeting of the Association for Computational Linguistics, ACL 2017 ; Conference date: 30-07-2017 Through 04-08-2017",

year = "2017",

doi = "10.18653/v1/P17-1134",

language = "English",

volume = "1",

pages = "1457--1469",

editor = "Regina Barzilay and Min-Yen Kan",

booktitle = "ACL 2017",

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

}

Malmasi, S , Dras, M , Johnson, M, Du, L & Wolska, M 2017, Unsupervised text segmentation based on native language characteristics. in R Barzilay & M-Y Kan (eds), ACL 2017: Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Long Papers). vol. 1, Association for Computational Linguistics (ACL), Stroudsburg, PA, pp. 1457-1469, 55th Annual Meeting of the Association for Computational Linguistics, ACL 2017, Vancouver, Canada, 30/07/17. https://doi.org/10.18653/v1/P17-1134

Unsupervised text segmentation based on native language characteristics. / Malmasi, Shervin ; Dras, Mark ; Johnson, Mark et al.
ACL 2017: Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Long Papers). ed. / Regina Barzilay; Min-Yen Kan. Vol. 1 Stroudsburg, PA: Association for Computational Linguistics (ACL), 2017. p. 1457-1469.

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

TY - GEN

T1 - Unsupervised text segmentation based on native language characteristics

AU - Malmasi, Shervin

AU - Dras, Mark

AU - Johnson, Mark

AU - Du, Lan

AU - Wolska, Magdalena

N1 - Copyright the Publisher. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

PY - 2017

Y1 - 2017

N2 - Most work on segmenting text does so on the basis of topic changes, but it can be of interest to segment by other, stylistically expressed characteristics such as change of authorship or native language. We propose a Bayesian unsupervised text segmentation approach to the latter. While baseline models achieve essentially random segmentation on our task, indicating its difficulty, a Bayesian model that incorporates appropriately compact language models and alternating asymmetric priors can achieve scores on the standard metrics around halfway to perfect segmentation.

AB - Most work on segmenting text does so on the basis of topic changes, but it can be of interest to segment by other, stylistically expressed characteristics such as change of authorship or native language. We propose a Bayesian unsupervised text segmentation approach to the latter. While baseline models achieve essentially random segmentation on our task, indicating its difficulty, a Bayesian model that incorporates appropriately compact language models and alternating asymmetric priors can achieve scores on the standard metrics around halfway to perfect segmentation.

UR - http://www.scopus.com/inward/record.url?scp=85040930774&partnerID=8YFLogxK

U2 - 10.18653/v1/P17-1134

DO - 10.18653/v1/P17-1134

M3 - Conference proceeding contribution

AN - SCOPUS:85040930774

VL - 1

SP - 1457

EP - 1469

BT - ACL 2017

A2 - Barzilay, Regina

A2 - Kan, Min-Yen

PB - Association for Computational Linguistics (ACL)

CY - Stroudsburg, PA

T2 - 55th Annual Meeting of the Association for Computational Linguistics, ACL 2017

Y2 - 30 July 2017 through 4 August 2017

ER -

Malmasi S , Dras M , Johnson M, Du L, Wolska M. Unsupervised text segmentation based on native language characteristics. In Barzilay R, Kan MY, editors, ACL 2017: Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Long Papers). Vol. 1. Stroudsburg, PA: Association for Computational Linguistics (ACL). 2017. p. 1457-1469 doi: 10.18653/v1/P17-1134

Unsupervised text segmentation based on native language characteristics

Abstract

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this