Generalized character-level spelling error correction

Noura Farra, Nadi Tomeh, Alla Rozovskaya, Nizar Habash

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present a generalized discriminative model for spelling error correction which targets character-level transformations. While operating at the character level, the model makes use of wordlevel and contextual information. In contrast to previous work, the proposed approach learns to correct a variety of error types without guidance of manually-selected constraints or language-specific features. We apply the model to correct errors in Egyptian Arabic dialect text, achieving 65% reduction in word error rate over the input baseline, and improving over the earlier state-of-the-art system.

Original languageEnglish (US)
Title of host publicationLong Papers
PublisherAssociation for Computational Linguistics (ACL)
Pages161-167
Number of pages7
Volume2
ISBN (Print)9781937284732
StatePublished - Jan 1 2014
Event52nd Annual Meeting of the Association for Computational Linguistics, ACL 2014 - Baltimore, MD, United States
Duration: Jun 22 2014Jun 27 2014

Other

Other52nd Annual Meeting of the Association for Computational Linguistics, ACL 2014
CountryUnited States
CityBaltimore, MD
Period6/22/146/27/14

Fingerprint

dialect
Spelling
Error Correction
language
Early States
Egyptians
Arabic Dialects
Contextual
Guidance
Language

ASJC Scopus subject areas

  • Language and Linguistics
  • Linguistics and Language

Cite this

Farra, N., Tomeh, N., Rozovskaya, A., & Habash, N. (2014). Generalized character-level spelling error correction. In Long Papers (Vol. 2, pp. 161-167). Association for Computational Linguistics (ACL).

Generalized character-level spelling error correction. / Farra, Noura; Tomeh, Nadi; Rozovskaya, Alla; Habash, Nizar.

Long Papers. Vol. 2 Association for Computational Linguistics (ACL), 2014. p. 161-167.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Farra, N, Tomeh, N, Rozovskaya, A & Habash, N 2014, Generalized character-level spelling error correction. in Long Papers. vol. 2, Association for Computational Linguistics (ACL), pp. 161-167, 52nd Annual Meeting of the Association for Computational Linguistics, ACL 2014, Baltimore, MD, United States, 6/22/14.
Farra N, Tomeh N, Rozovskaya A, Habash N. Generalized character-level spelling error correction. In Long Papers. Vol. 2. Association for Computational Linguistics (ACL). 2014. p. 161-167
Farra, Noura ; Tomeh, Nadi ; Rozovskaya, Alla ; Habash, Nizar. / Generalized character-level spelling error correction. Long Papers. Vol. 2 Association for Computational Linguistics (ACL), 2014. pp. 161-167
@inproceedings{b463b399e1994700a9530fc95ced511c,
title = "Generalized character-level spelling error correction",
abstract = "We present a generalized discriminative model for spelling error correction which targets character-level transformations. While operating at the character level, the model makes use of wordlevel and contextual information. In contrast to previous work, the proposed approach learns to correct a variety of error types without guidance of manually-selected constraints or language-specific features. We apply the model to correct errors in Egyptian Arabic dialect text, achieving 65{\%} reduction in word error rate over the input baseline, and improving over the earlier state-of-the-art system.",
author = "Noura Farra and Nadi Tomeh and Alla Rozovskaya and Nizar Habash",
year = "2014",
month = "1",
day = "1",
language = "English (US)",
isbn = "9781937284732",
volume = "2",
pages = "161--167",
booktitle = "Long Papers",
publisher = "Association for Computational Linguistics (ACL)",

}

TY - GEN

T1 - Generalized character-level spelling error correction

AU - Farra, Noura

AU - Tomeh, Nadi

AU - Rozovskaya, Alla

AU - Habash, Nizar

PY - 2014/1/1

Y1 - 2014/1/1

N2 - We present a generalized discriminative model for spelling error correction which targets character-level transformations. While operating at the character level, the model makes use of wordlevel and contextual information. In contrast to previous work, the proposed approach learns to correct a variety of error types without guidance of manually-selected constraints or language-specific features. We apply the model to correct errors in Egyptian Arabic dialect text, achieving 65% reduction in word error rate over the input baseline, and improving over the earlier state-of-the-art system.

AB - We present a generalized discriminative model for spelling error correction which targets character-level transformations. While operating at the character level, the model makes use of wordlevel and contextual information. In contrast to previous work, the proposed approach learns to correct a variety of error types without guidance of manually-selected constraints or language-specific features. We apply the model to correct errors in Egyptian Arabic dialect text, achieving 65% reduction in word error rate over the input baseline, and improving over the earlier state-of-the-art system.

UR - http://www.scopus.com/inward/record.url?scp=84906922684&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84906922684&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84906922684

SN - 9781937284732

VL - 2

SP - 161

EP - 167

BT - Long Papers

PB - Association for Computational Linguistics (ACL)

ER -