A Comparison of Neural Networks Architectures for Diacritics Restoration

There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

Abstract

Neural networks are widely used for the task of diacritics restoration last years. Authors use different architectures of neural network for selected languages. In this paper, we demonstrated that an architecture should be selected according to a language in hand. It also depends on a task one states: low and full resourced languages could use different architectures. We demonstrated that common used accuracy metric should be changed in this task to precision and recall due to the heavy unbalanced nature of the input data. The paper contains results for seven languages: Croatian, Slovak, Romanian, French, German, Latvian, and Turkish.

Related collections

Most cited references 2

Record: found
Abstract: not found
Conference Proceedings: not found

Automatic restoration of diacritics for Igbo language

I Ezeani, M Hepple, I. Onyenwe (2016)

0 comments Cited 1 times – based on 0 reviews

Bookmark

Record: found
Abstract: not found
Book: not found

Comparison of corpus-based techniques for restoring accents in Spanish and French text

D Yarowsky (1999)

0 comments Cited 1 times – based on 0 reviews

Bookmark

All references

Author and article information

Contributors

Wil M. P. van der Aalst: wvdaalst@pads.rwth-aachen.de

Vladimir Batagelj: vladimir.batagelj@fmf.uni-lj.si

Alexey Buzmakov: AVBuzmakov@hse.ru

Dmitry I. Ignatov: dmitrii.ignatov@gmail.com

Anna Kalenkova: anna.kalenkova@unimelb.edu.au

Michael Khachay: mkhachay@imm.uran.ru

Olessia Koltsova: ekoltsova@hse.ru

Andrey Kutuzov: andreku@ifi.uio.no

Sergei O. Kuznetsov: skuznetsov@hse.ru

Irina A. Lomazova: ilomazova@hse.ru

Natalia Loukachevitch: louk_nat@mail.ru

Ilya Makarov: iamakarov@hse.ru

Amedeo Napoli: amedeo.napoli@loria.fr

Alexander Panchenko: panchenkoalexander@gmail.com

Panos M. Pardalos: pardalos@ufl.edu

Marcello Pelillo: pelillo@dsi.unive.it

Andrey V. Savchenko: avsavchenko@hse.ru

Elena Tutubalina: elvtutubalina@kpfu.ru

Eduard Klyshinsky:

ORCID: http://orcid.org/0000-0002-4020-488X

klyshinsky@mail.ru

Olesya Karpik:

ORCID: http://orcid.org/0000-0002-0477-1502

parlak@mail.ru

Alexander Bondarenko:

ORCID: http://orcid.org/0000-0003-4765-6034

cod@fgosniias.ru

Journal

Journal ID (publisher-id): 978-3-030-71214-3

Journal ID (doi): 10.1007/978-3-030-71214-3

Journal ID (nlm-ta): Recent Trends in Analysis of Images, Social Networks and Texts

Title: Recent Trends in Analysis of Images, Social Networks and Texts

Subtitle: 9th International Conference, AIST 2020, Skolkovo, Moscow, Russia, October 15–16, 2020 Revised Supplementary Proceedings

ISBN (Print): 978-3-030-71213-6

ISBN (Electronic): 978-3-030-71214-3

Publication date (Electronic): 20 February 2021

Publication date Collection: 2021

Volume: 1357

Pages: 242-253

Affiliations

[5 ]GRID grid.1957.a, ISNI 0000 0001 0728 696X, RWTH Aachen University, ; Aachen, Germany

[6 ]GRID grid.8954.0, ISNI 0000 0001 0721 6013, University of Ljubljana, ; Ljubljana, Slovenia

[7 ]GRID grid.410682.9, ISNI 0000 0004 0578 2005, National Research University Higher School of Economics, ; Perm, Russia

[8 ]GRID grid.410682.9, ISNI 0000 0004 0578 2005, National Research University Higher School of Economics, ; Moscow, Russia

[9 ]GRID grid.1008.9, ISNI 0000 0001 2179 088X, University of Melbourne, ; Melbourne, VIC Australia

[10 ]Krasovskii Institute of Mathematics and Mechanics of RAS, Ekaterinburg, Russia

[11 ]GRID grid.410682.9, ISNI 0000 0004 0578 2005, National Research University Higher School of Economics, ; Saint-Petersburg, Russia

[12 ]GRID grid.5510.1, ISNI 0000 0004 1936 8921, University of Oslo, ; Oslo, Norway

[13 ]GRID grid.410682.9, ISNI 0000 0004 0578 2005, National Research University Higher School of Economics, ; Moscow, Russia

[14 ]GRID grid.410682.9, ISNI 0000 0004 0578 2005, National Research University Higher School of Economics, ; Moscow, Russia

[15 ]GRID grid.14476.30, ISNI 0000 0001 2342 9668, Lomonosov Moscow State University, ; Moscow, Russia

[16 ]GRID grid.410682.9, ISNI 0000 0004 0578 2005, National Research University Higher School of Economics, ; Moscow, Russia

[17 ]GRID grid.462764.5, ISNI 0000 0001 2179 5429, LORIA, ; Vandœuvre-lès-Nancy, France

[18 ]GRID grid.454320.4, ISNI 0000 0004 0555 3608, Skolkovo Institute of Science and Technology, ; Moscow, Russia

[19 ]GRID grid.15276.37, ISNI 0000 0004 1936 8091, University of Florida, ; Gainesville, FL USA

[20 ]GRID grid.7240.1, ISNI 0000 0004 1763 0578, Università Ca’ Foscari Venezia, ; Venezia, Italy

[21 ]GRID grid.410682.9, ISNI 0000 0004 0578 2005, National Research University Higher School of Economics, ; Nizhny Novgorod, Russia

[22 ]GRID grid.77268.3c, ISNI 0000 0004 0543 9688, Kazan Federal University, ; Kazan, Russia

[23 ]GRID grid.410682.9, ISNI 0000 0004 0578 2005, National Research University Higher School of Economics, ; S. Basmannaya Street, 21, Moscow, 105066 Russia

[24 ]GRID grid.435669.b, ISNI 0000 0001 0673 1283, Keldysh Institute of Applied Mathematics, ; Miusskaya sq., 4, Moscow, 125047 Russia

[25 ]GRID grid.459978.e, ISNI 0000 0001 2108 5662, State Research Institute of Aviation Systems, ; Viktorenko Street, 7, Moscow, 125319 Russia

Author information

Eduard Klyshinsky http://orcid.org/0000-0002-4020-488X

Olesya Karpik http://orcid.org/0000-0002-0477-1502

Alexander Bondarenko http://orcid.org/0000-0003-4765-6034

Article

Publisher ID: 20

DOI: 10.1007/978-3-030-71214-3_20

PMC ID: 7988423

SO-VID: ff03cdb3-53d1-4eb2-9014-a0000ef37262

License:

This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.

History

Custom metadata

Keywords: diacritics restoration,neural networks,out-of-vocabulary words

Data availability:

Keywords: diacritics restoration, neural networks, out-of-vocabulary words

A Comparison of Neural Networks Architectures for Diacritics Restoration

Read this article at

Abstract

Related collections

Data-Driven Civil Engineering

Most cited references 2

Automatic restoration of diacritics for Igbo language

Comparison of corpus-based techniques for restoring accents in Spanish and French text

Author and article information

Contributors

Journal

Affiliations

Author information

Article

History

Categories

Custom metadata

Comments

Comment on this article

Similar content 155

Most referenced authors 4