Data-Driven Spelling Correction using Weighted Finite-State Methods

Miikka Silfverberg; Pekka Kauppinen; Krister Linden

doi:10.18653/v1/w16-2406

Data-Driven Spelling Correction using Weighted Finite-State Methods

Miikka Silfverberg, Pekka Kauppinen, Krister Linden

Språkteknologi

Forskningsoutput: Kapitel i bok/rapport/konferenshandling › Konferensbidrag › Vetenskaplig › Peer review

Sammanfattning

This paper presents two systems for spelling correction formulated as a sequence
labeling task. One of the systems is an unstructured classifier and the other
one is structured. Both systems are implemented using weighted finite-state methods. The structured system delivers stateof-the-art results on the task of tweet normalization when compared with the recent AliSeTra system introduced by Eger et al. (2016) even though the system presented in the paper is simpler than AliSeTra because it does not include a model for input segmentation. In addition to experiments on tweet normalization, we present experiments on OCR post-processing using an Early Modern Finnish corpus of OCR processed newspaper text.

Originalspråk	engelska
Titel på värdpublikation	The 54th Annual Meeting of the Association for Computational Linguistics : Proceedings of the SIGFSM Workshop on Statistical NLP and Weighted Automata
Antal sidor	9
Utgivningsort	Stroudsburg, PA
Förlag	ACL
Utgivningsdatum	12 aug. 2016
Sidor	51-59
ISBN (tryckt)	978-1-945626-13-5, 1-945626-13-5
DOI	https://doi.org/10.18653/v1/w16-2406
Status	Publicerad - 12 aug. 2016
MoE-publikationstyp	A4 Artikel i en konferenspublikation
Evenemang	Annual Meeting of the Association for Computational Linguistics - Berlin, Tyskland Varaktighet: 12 okt. 2016 → 12 okt. 2016 Konferensnummer: 54

Vetenskapsgrenar

6121 Språkvetenskaper
113 Data- och informationsvetenskap

Åtkomst av dokument

10.18653/v1/w16-2406Licens: CC BY

W16-2406Slutlig publicerad version, 173 KBLicens: CC BY

https://aclweb.org/anthology/W/W16/W16-2406.pdfLicens: CC BY

CLARIN - Finländska språkresurser i gemensamt bruk
Linden, K. (Chef)
Avdelningen för digital humaniora
Utrustning/facilitet: Coordination office

Supervisor of doctoral thesis in Language Technology / Miikka Silfverberg
Linden, B. K. J. (Handledare) & Yli-Jyrä, A. (Handledare)
1 jan. 2010 → 22 okt. 2016
Aktivitet: Examinationstyper › Handledare eller bihandledare av doktorsavhandling

Citera det här

@inproceedings{2eedf63901c74f73816d6330b18c7ba8,

title = "Data-Driven Spelling Correction using Weighted Finite-State Methods",

abstract = "This paper presents two systems for spelling correction formulated as a sequencelabeling task. One of the systems is an unstructured classifier and the otherone is structured. Both systems are implemented using weighted finite-state methods. The structured system delivers stateof-the-art results on the task of tweet normalization when compared with the recent AliSeTra system introduced by Eger et al. (2016) even though the system presented in the paper is simpler than AliSeTra because it does not include a model for input segmentation. In addition to experiments on tweet normalization, we present experiments on OCR post-processing using an Early Modern Finnish corpus of OCR processed newspaper text.",

keywords = "6121 Languages, 113 Computer and information sciences",

author = "Miikka Silfverberg and Pekka Kauppinen and Krister Linden",

year = "2016",

month = aug,

day = "12",

doi = "10.18653/v1/w16-2406",

language = "English",

isbn = "978-1-945626-13-5",

pages = "51--59",

booktitle = "The 54th Annual Meeting of the Association for Computational Linguistics",

publisher = "ACL",

address = "United States",

note = "Annual Meeting of the Association for Computational Linguistics ; Conference date: 12-10-2016 Through 12-10-2016",

}

Silfverberg, M, Kauppinen, P & Linden, K 2016, Data-Driven Spelling Correction using Weighted Finite-State Methods. i The 54th Annual Meeting of the Association for Computational Linguistics: Proceedings of the SIGFSM Workshop on Statistical NLP and Weighted Automata. ACL, Stroudsburg, PA, s. 51-59, Annual Meeting of the Association for Computational Linguistics, Berlin, Tyskland, 12/10/2016. https://doi.org/10.18653/v1/w16-2406

Data-Driven Spelling Correction using Weighted Finite-State Methods. / Silfverberg, Miikka; Kauppinen, Pekka; Linden, Krister.
The 54th Annual Meeting of the Association for Computational Linguistics: Proceedings of the SIGFSM Workshop on Statistical NLP and Weighted Automata. Stroudsburg, PA: ACL, 2016. s. 51-59.

Forskningsoutput: Kapitel i bok/rapport/konferenshandling › Konferensbidrag › Vetenskaplig › Peer review

TY - GEN

T1 - Data-Driven Spelling Correction using Weighted Finite-State Methods

AU - Silfverberg, Miikka

AU - Kauppinen, Pekka

AU - Linden, Krister

N1 - Conference code: 54

PY - 2016/8/12

Y1 - 2016/8/12

N2 - This paper presents two systems for spelling correction formulated as a sequencelabeling task. One of the systems is an unstructured classifier and the otherone is structured. Both systems are implemented using weighted finite-state methods. The structured system delivers stateof-the-art results on the task of tweet normalization when compared with the recent AliSeTra system introduced by Eger et al. (2016) even though the system presented in the paper is simpler than AliSeTra because it does not include a model for input segmentation. In addition to experiments on tweet normalization, we present experiments on OCR post-processing using an Early Modern Finnish corpus of OCR processed newspaper text.

AB - This paper presents two systems for spelling correction formulated as a sequencelabeling task. One of the systems is an unstructured classifier and the otherone is structured. Both systems are implemented using weighted finite-state methods. The structured system delivers stateof-the-art results on the task of tweet normalization when compared with the recent AliSeTra system introduced by Eger et al. (2016) even though the system presented in the paper is simpler than AliSeTra because it does not include a model for input segmentation. In addition to experiments on tweet normalization, we present experiments on OCR post-processing using an Early Modern Finnish corpus of OCR processed newspaper text.

KW - 6121 Languages

KW - 113 Computer and information sciences

U2 - 10.18653/v1/w16-2406

DO - 10.18653/v1/w16-2406

M3 - Conference contribution

SN - 978-1-945626-13-5

SN - 1-945626-13-5

SP - 51

EP - 59

BT - The 54th Annual Meeting of the Association for Computational Linguistics

PB - ACL

CY - Stroudsburg, PA

T2 - Annual Meeting of the Association for Computational Linguistics

Y2 - 12 October 2016 through 12 October 2016

ER -

Data-Driven Spelling Correction using Weighted Finite-State Methods

Sammanfattning

Vetenskapsgrenar

Åtkomst av dokument

Utrustning

CLARIN - Finländska språkresurser i gemensamt bruk

Aktiviteter

Supervisor of doctoral thesis in Language Technology / Miikka Silfverberg

Citera det här