short-paper

Neural Machine Translation with BERT for Post-OCR Error Detection and Correction

Authors:

Thi Tuyet Hai Nguyen,

Adam Jatowt,

Nhu-Van Nguyen,

Mickael Coustaty,

Antoine DoucetAuthors Info & Claims

JCDL '20: Proceedings of the ACM/IEEE Joint Conference on Digital Libraries in 2020

Pages 333 - 336

https://doi.org/10.1145/3383583.3398605

Published: 01 August 2020 Publication History

Get Access

Abstract

The quality of OCR has a direct impact on information access, and an indirect impact on the performance of natural language processing applications, making fine-grained (e.g., semantic) information access even harder. This work proposes a novel post-OCR approach based on a contextual language model and neural machine translation, aiming to improve the quality of OCRed text by detecting and rectifying erroneous tokens. This new technique obtains results comparable to the best-performing approaches on English datasets of the competition on post-OCR text correction in ICDAR 2017/2019.

Supplementary Material

MP4 File (3383583.3398605.mp4)

I would like to present our post-OCR correction approach which applies neural machine translation and contextual language model BERT.

Download
10.75 MB

References

[1]

Chantal Amrhein and Simon Clematide. 2018. Supervised OCR Error Detection and Correction Using Statistical and Neural Machine Translation Methods. Journal for Language Technology and Computational Linguistics (2018).

Google Scholar

[2]

Guillaume Chiron, Antoine Doucet, Mickael Coustaty, and Jean-Philippe Moreux. 2017. ICDAR2017 competition on post-OCR text correction. In 14th IAPR International Conference on Document Analysis and Recognition. IEEE, 1423--1428.

Google Scholar

[3]

Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2018. Bert: Pretraining of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805 (2018).

Google Scholar

[4]

John Evershed and Kent Fitch. 2014. Correcting noisy OCR: Context beats confusion. In Proceedings of the First International Conference on Digital Access to Textual Cultural Heritage. ACM, 45--51.

Digital Library

Google Scholar

[5]

Ido Kissos and Nachum Dershowitz. 2016. OCR error correction using character correction and feature-based word classification. In Document Analysis Systems (DAS), 2016 12th IAPR Workshop on. IEEE, 198--203.

Crossref

Google Scholar

[6]

Guillaume Klein, Yoon Kim, Yuntian Deng, Jean Senellart, and AlexanderRush. 2017. OpenNMT: Open-Source Toolkit for Neural Machine Translation. In Proceedings of ACL 2017, System Demonstrations. 67--72.

Crossref

Google Scholar

[7]

Bryan McCann, James Bradbury, Caiming Xiong, and Richard Socher. 2017. Learned in translation: Contextualized word vectors. In Advances in Neural Information Processing Systems. 6294--6305.

Google Scholar

[8]

Thi-Tuyet-Hai Nguyen, Mickael Coustaty, Antoine Doucet, Adam Jatowt, and Nhu-Van Nguyen. 2018. Adaptive Edit-Distance and Regression Approach for Post-OCR Text Correction. In 20th International Conference on Asia-Pacific Digital Libraries, ICADL 2018. 278--289.

Google Scholar

[9]

Thi-Tuyet-Hai Nguyen, Adam Jatowt, Mickael Coustaty, Nhu-Van Nguyen, and Antoine Doucet. 2019. Deep Statistical Analysis of OCR Errors for Effective Post-OCR Processing. In 19th ACM/IEEE Joint Conf. on Digital Libraries. 29--38.

Google Scholar

[10]

Thi-Tuyet-Hai Nguyen, Adam Jatowt, Mickael Coustaty, Nhu-Van Nguyen, and Antoine Doucet. 2019. Post-OCR Error Detection by Generating Plausible Candidates. In 2019 International Conference on Document Analysis and Recognition, ICDAR 2019, Sydney, Australia, September 20--25, 2019. IEEE, 876--881.

Crossref

Google Scholar

[11]

Christophe Rigaud, Antoine Doucet, Mickael Coustaty, and Jean-Philippe Moreux. 2019. ICDAR 2019 Competition on Post-OCR Text Correction. In 15th IAPR International Conference on Document Analysis and Recognition, ICDAR 2019.

Google Scholar

[12]

Sarah Schulz and Jonas Kuhn. 2017. Multi-modular domain-tailored OCR post-correction. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. 2716--2726.

Crossref

Google Scholar

[13]

Myriam C Traub, Jacco Van Ossenbruggen, and Lynda Hardman. 2015. Impact analysis of OCR quality on research tasks in digital archives. In International Conference on Theory and Practice of Digital Libraries. Springer, 252--263.

Crossref

Google Scholar

[14]

Yonghui Wu, Mike Schuster, Zhifeng Chen, et al. 2016. Google's neural machine translation system: Bridging the gap between human and machine translation. arXiv preprint arXiv:1609.08144 (2016).

Google Scholar

Cited By

View all

Petkovic LKoudoro-Parfait CDesmarest MLejeune G(2025)Quelle solution pour améliorer les performances de la reconnaissance d’entités nommées sur des données bruitées, corriger l’entrée ou filtrer la sortie ?Corpus10.4000/1364s26Online publication date: 2025
https://doi.org/10.4000/1364s
Shen WLi HJin YWu C(2025)End-to-End Information Extraction from Courier Order Images Using a Neural Network Model with Feature EnhancementApplied Sciences10.3390/app1502069815:2(698)Online publication date: 12-Jan-2025
https://doi.org/10.3390/app15020698
Zhang MCui QLi WYu WChen LLi WZhu CLü Y(2025)Augmented dialectal speech recognition for AI-based neuropsychological scale assessment in Alzheimer’s diseaseBiomedical Signal Processing and Control10.1016/j.bspc.2024.10682199(106821)Online publication date: Jan-2025
https://doi.org/10.1016/j.bspc.2024.106821
Show More Cited By

Index Terms

Neural Machine Translation with BERT for Post-OCR Error Detection and Correction
1. Applied computing
  1. Document management and text processing
2. Information systems
  1. Information systems applications
    1. Digital libraries and archives

Recommendations

Transformer-Based Neural Machine Translation for Post-OCR Error Correction in Cursive Text
Document Analysis and Recognition – ICDAR 2023 Workshops
Abstract
This study investigates transformer-based Neural Machine Translation (NMT) for post-processing of noisy Optical Character Recognition (OCR) output. While recognition engines have matured significantly for most languages, the problem still remains ...
Post-editing neural machine translation versus phrase-based machine translation for English---Chinese

This paper aims to shed light on the post-editing process of the recently-introduced neural machine translation (NMT) paradigm. Using simple and more complex texts, we first evaluate the output quality from English to Chinese phrase-based statistical (...
A study of BERT for context-aware neural machine translation
Abstract
Context-aware neural machine translation (NMT), which targets at translating sentences with contextual information, has attracted much attention recently. A key problem for context-aware NMT is to effectively encode and aggregate the contextual ...

Comments

Information & Contributors

Information

Published In

JCDL '20: Proceedings of the ACM/IEEE Joint Conference on Digital Libraries in 2020

August 2020

611 pages

ISBN:9781450375856

DOI:10.1145/3383583

General Chairs:
Ruhua Huang
Wuhan University, China
,
Dan Wu
Wuhan University, China
,
Gary Marchionini
University of North Carolina at Chapel Hill, USA
,
Program Chairs:
Daqing He
University of Pittsburgh, USA
,
Sally Jo Cunningham
University of Waikato, New Zealand
,
Preben Hansen
Stockholm University, Sweden

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 01 August 2020

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Short-paper

Funding Sources

H2020

Conference

JCDL '20

Sponsor:

JCDL '20: The ACM/IEEE Joint Conference on Digital Libraries in 2020

August 1 - 5, 2020

Virtual Event, China

Acceptance Rates

Overall Acceptance Rate 415 of 1,482 submissions, 28%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

23
Total Citations
View Citations
776
Total Downloads

Downloads (Last 12 months)93
Downloads (Last 6 weeks)8

Reflects downloads up to 16 Feb 2025

Other Metrics

View Author Metrics

Citations

Cited By

View all

Petkovic LKoudoro-Parfait CDesmarest MLejeune G(2025)Quelle solution pour améliorer les performances de la reconnaissance d’entités nommées sur des données bruitées, corriger l’entrée ou filtrer la sortie ?Corpus10.4000/1364s26Online publication date: 2025
https://doi.org/10.4000/1364s
Shen WLi HJin YWu C(2025)End-to-End Information Extraction from Courier Order Images Using a Neural Network Model with Feature EnhancementApplied Sciences10.3390/app1502069815:2(698)Online publication date: 12-Jan-2025
https://doi.org/10.3390/app15020698
Zhang MCui QLi WYu WChen LLi WZhu CLü Y(2025)Augmented dialectal speech recognition for AI-based neuropsychological scale assessment in Alzheimer’s diseaseBiomedical Signal Processing and Control10.1016/j.bspc.2024.10682199(106821)Online publication date: Jan-2025
https://doi.org/10.1016/j.bspc.2024.106821
Tavakoli MChandra RTian FBravo C(2025)Multi-modal deep learning for credit rating prediction using text and numerical data streamsApplied Soft Computing10.1016/j.asoc.2025.112771171(112771)Online publication date: Mar-2025
https://doi.org/10.1016/j.asoc.2025.112771
Villa-García PAlonso-Calvo RGarcía-Remesal M(2024)End-to-end entity extraction from OCRed texts using summarization modelsNeural Computing and Applications10.1007/s00521-024-10422-936:35(22347-22363)Online publication date: 1-Dec-2024
https://dl.acm.org/doi/10.1007/s00521-024-10422-9
Hemmer ACoustaty MBartolo NOgier J(2024)Confidence-Aware Document OCR Error DetectionDocument Analysis Systems10.1007/978-3-031-70442-0_13(213-228)Online publication date: 11-Sep-2024
https://doi.org/10.1007/978-3-031-70442-0_13
Zening LJiapeng WLianwen J(2023)Visual information extraction deep learning method：a critical reviewJournal of Image and Graphics10.11834/jig.22090428:8(2276-2297)Online publication date: 2023
https://doi.org/10.11834/jig.220904
Prakash YSunkara MLee HJayarathna SAshok V(2023)AutoDesc: Facilitating Convenient Perusal of Web Data Items for Blind UsersProceedings of the 28th International Conference on Intelligent User Interfaces10.1145/3581641.3584049(32-45)Online publication date: 27-Mar-2023
https://dl.acm.org/doi/10.1145/3581641.3584049
Guan YPang ZDing YTian L(2023)Text error correction after text recognition based on MacBERT4CSCSixth International Conference on Advanced Electronic Materials, Computers, and Software Engineering (AEMCSE 2023)10.1117/12.3004939(97)Online publication date: 16-Aug-2023
https://doi.org/10.1117/12.3004939
Yokoyama MShimada K(2023)Detecting speech recognition errors using topic information and BERT2023 14th IIAI International Congress on Advanced Applied Informatics (IIAI-AAI)10.1109/IIAI-AAI59060.2023.00084(400-405)Online publication date: 8-Jul-2023
https://doi.org/10.1109/IIAI-AAI59060.2023.00084
Show More Cited By

View Options

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

Abstract

Supplementary Material

References

Cited By

Index Terms

Recommendations

Transformer-Based Neural Machine Translation for Post-OCR Error Correction in Cursive Text

Post-editing neural machine translation versus phrase-based machine translation for English---Chinese

A study of BERT for context-aware neural machine translation

Comments

Information

Published In

Sponsors

Publisher

Publication History

Permissions

Check for updates

Author Tags

Qualifiers

Funding Sources

Conference

Acceptance Rates

Contributors

Other Metrics

Bibliometrics

Article Metrics

Other Metrics

Citations

Cited By

Login options

Full Access

View options

PDF

eReader

Share

Share this Publication link

Share on social media

Affiliations