The CLIN27 Shared Task: Translating Historical Text to Contemporary Language for Improving Automatic Linguistic Annotation

Erik Tjong Kim Sang; Marcel Bollman; Remko Boschker; Francisco Casacuberta; F.M. Dietz; Stefanie Dipper; Miguel Domingo; Rob van der Goot; J.M. van Koppen; Nikola Ljubešić; Robert Östling; Florian Petran; Eva Pettersson; Yves Scherrer; M.P. Schraagen; Leen Sevens; Jörg Tiedeman; Tom Vanallemeersch; K. Zervanou

The CLIN27 Shared Task: Translating Historical Text to Contemporary Language for Improving Automatic Linguistic Annotation

Erik Tjong Kim Sang, Marcel Bollman, Remko Boschker, Francisco Casacuberta, F.M. Dietz, Stefanie Dipper, Miguel Domingo, Rob van der Goot, J.M. van Koppen, Nikola Ljubešić, Robert Östling, Florian Petran, Eva Pettersson, Yves Scherrer, M.P. Schraagen, Leen Sevens, Jörg Tiedeman, Tom Vanallemeersch, K. Zervanou

ICON - Early Modern Literature

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

The CLIN27 shared task evaluates the effect of translating historical text to modern text with the goal of improving the quality of the output of contemporary natural language processing tools applied to the text. We focus on improving part-of-speech tagging analysis of seventeenth-century Dutch. Eight teams took part in the shared task. The best results were obtained by teams employing character-based machine translation. The best system obtained an error reduction of 51% in comparison with the baseline of tagging unmodified text. This is close to the error reduction obtained by human translation (57%).

Original language	English
Pages (from-to)	53-64
Journal	Computational Linguistics in the Netherlands Journal
Volume	7
Publication status	Published - Dec 2017

Keywords

historical text
text normalization
neural networks
machine translation
dutch language

Access to Document

04.clin27-shared-taskFinal published version, 252 KB

https://clinjournal.org/clinj/article/view/68

Cite this

Tjong Kim Sang, E., Bollman, M., Boschker, R., Casacuberta, F., Dietz, F. M., Dipper, S., Domingo, M., van der Goot, R., van Koppen, J. M., Ljubešić, N., Östling, R., Petran, F., Pettersson, E., Scherrer, Y., Schraagen, M. P., Sevens, L., Tiedeman, J., Vanallemeersch, T., & Zervanou, K. (2017). The CLIN27 Shared Task: Translating Historical Text to Contemporary Language for Improving Automatic Linguistic Annotation. Computational Linguistics in the Netherlands Journal, 7, 53-64. https://clinjournal.org/clinj/article/view/68

@article{5909bd807f424dbfbd66d5f2b42e2c5b,

title = "The CLIN27 Shared Task: Translating Historical Text to Contemporary Language for Improving Automatic Linguistic Annotation",

abstract = "The CLIN27 shared task evaluates the effect of translating historical text to modern text with the goal of improving the quality of the output of contemporary natural language processing tools applied to the text. We focus on improving part-of-speech tagging analysis of seventeenth-century Dutch. Eight teams took part in the shared task. The best results were obtained by teams employing character-based machine translation. The best system obtained an error reduction of 51\% in comparison with the baseline of tagging unmodified text. This is close to the error reduction obtained by human translation (57\%).",

keywords = "historical text, text normalization, neural networks, machine translation, dutch language",

author = "\{Tjong Kim Sang\}, Erik and Marcel Bollman and Remko Boschker and Francisco Casacuberta and F.M. Dietz and Stefanie Dipper and Miguel Domingo and \{van der Goot\}, Rob and \{van Koppen\}, J.M. and Nikola Ljube{\v s}i{\'c} and Robert {\"O}stling and Florian Petran and Eva Pettersson and Yves Scherrer and M.P. Schraagen and Leen Sevens and J{\"o}rg Tiedeman and Tom Vanallemeersch and K. Zervanou",

year = "2017",

month = dec,

language = "English",

volume = "7",

pages = "53--64",

journal = "Computational Linguistics in the Netherlands Journal",

issn = "2211-4009",

publisher = "Computational Linguistics in the Netherlands",

}

Tjong Kim Sang, E, Bollman, M, Boschker, R, Casacuberta, F, Dietz, FM, Dipper, S, Domingo, M, van der Goot, R, van Koppen, JM, Ljubešić, N, Östling, R, Petran, F, Pettersson, E, Scherrer, Y, Schraagen, MP, Sevens, L, Tiedeman, J, Vanallemeersch, T & Zervanou, K 2017, 'The CLIN27 Shared Task: Translating Historical Text to Contemporary Language for Improving Automatic Linguistic Annotation', Computational Linguistics in the Netherlands Journal, vol. 7, pp. 53-64. <https://clinjournal.org/clinj/article/view/68>

TY - JOUR

T1 - The CLIN27 Shared Task

T2 - Translating Historical Text to Contemporary Language for Improving Automatic Linguistic Annotation

AU - Tjong Kim Sang, Erik

AU - Bollman, Marcel

AU - Boschker, Remko

AU - Casacuberta, Francisco

AU - Dietz, F.M.

AU - Dipper, Stefanie

AU - Domingo, Miguel

AU - van der Goot, Rob

AU - van Koppen, J.M.

AU - Ljubešić, Nikola

AU - Östling, Robert

AU - Petran, Florian

AU - Pettersson, Eva

AU - Scherrer, Yves

AU - Schraagen, M.P.

AU - Sevens, Leen

AU - Tiedeman, Jörg

AU - Vanallemeersch, Tom

AU - Zervanou, K.

PY - 2017/12

Y1 - 2017/12

N2 - The CLIN27 shared task evaluates the effect of translating historical text to modern text with the goal of improving the quality of the output of contemporary natural language processing tools applied to the text. We focus on improving part-of-speech tagging analysis of seventeenth-century Dutch. Eight teams took part in the shared task. The best results were obtained by teams employing character-based machine translation. The best system obtained an error reduction of 51% in comparison with the baseline of tagging unmodified text. This is close to the error reduction obtained by human translation (57%).

AB - The CLIN27 shared task evaluates the effect of translating historical text to modern text with the goal of improving the quality of the output of contemporary natural language processing tools applied to the text. We focus on improving part-of-speech tagging analysis of seventeenth-century Dutch. Eight teams took part in the shared task. The best results were obtained by teams employing character-based machine translation. The best system obtained an error reduction of 51% in comparison with the baseline of tagging unmodified text. This is close to the error reduction obtained by human translation (57%).

KW - historical text

KW - text normalization

KW - neural networks

KW - machine translation

KW - dutch language

M3 - Article

SN - 2211-4009

VL - 7

SP - 53

EP - 64

JO - Computational Linguistics in the Netherlands Journal

JF - Computational Linguistics in the Netherlands Journal

ER -

The CLIN27 Shared Task: Translating Historical Text to Contemporary Language for Improving Automatic Linguistic Annotation

Abstract

Keywords

Access to Document

Fingerprint

Cite this