<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lyu-etal-2021-neural">
    <titleInfo>
        <title>Neural OCR Post-Hoc Correction of Historical Corpora</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Lijun</namePart>
        <namePart type="family">Lyu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Maria</namePart>
        <namePart type="family">Koutraki</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Martin</namePart>
        <namePart type="family">Krickl</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Besnik</namePart>
        <namePart type="family">Fetahu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2021</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <genre authority="bibutilsgt">journal article</genre>
    <relatedItem type="host">
        <titleInfo>
            <title>Transactions of the Association for Computational Linguistics</title>
        </titleInfo>
        <originInfo>
            <issuance>continuing</issuance>
            <publisher>MIT Press</publisher>
            <place>
                <placeTerm type="text">Cambridge, MA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">periodical</genre>
        <genre authority="bibutilsgt">academic journal</genre>
    </relatedItem>
    <abstract>Optical character recognition (OCR) is crucial for a deeper access to historical collections. OCR needs to account for orthographic variations, typefaces, or language evolution (i.e., new letters, word spellings), as the main source of character, word, or word segmentation transcription errors. For digital corpora of historical prints, the errors are further exacerbated due to low scan quality and lack of language standardization. For the task of OCR post-hoc correction, we propose a neural approach based on a combination of recurrent (RNN) and deep convolutional network (ConvNet) to correct OCR transcription errors. At character level we flexibly capture errors, and decode the corrected output based on a novel attention mechanism. Accounting for the input and output similarity, we propose a new loss function that rewards the model’s correcting behavior. Evaluation on a historical book corpus in German language shows that our models are robust in capturing diverse OCR transcription errors and reduce the word error rate of 32.3% by more than 89%.</abstract>
    <identifier type="citekey">lyu-etal-2021-neural</identifier>
    <identifier type="doi">10.1162/tacl_a_00379</identifier>
    <location>
        <url>https://aclanthology.org/2021.tacl-1.29</url>
    </location>
    <part>
        <date>2021</date>
        <detail type="volume"><number>9</number></detail>
        <extent unit="page">
            <start>479</start>
            <end>493</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Journal Article
%T Neural OCR Post-Hoc Correction of Historical Corpora
%A Lyu, Lijun
%A Koutraki, Maria
%A Krickl, Martin
%A Fetahu, Besnik
%J Transactions of the Association for Computational Linguistics
%D 2021
%V 9
%I MIT Press
%C Cambridge, MA
%F lyu-etal-2021-neural
%X Optical character recognition (OCR) is crucial for a deeper access to historical collections. OCR needs to account for orthographic variations, typefaces, or language evolution (i.e., new letters, word spellings), as the main source of character, word, or word segmentation transcription errors. For digital corpora of historical prints, the errors are further exacerbated due to low scan quality and lack of language standardization. For the task of OCR post-hoc correction, we propose a neural approach based on a combination of recurrent (RNN) and deep convolutional network (ConvNet) to correct OCR transcription errors. At character level we flexibly capture errors, and decode the corrected output based on a novel attention mechanism. Accounting for the input and output similarity, we propose a new loss function that rewards the model’s correcting behavior. Evaluation on a historical book corpus in German language shows that our models are robust in capturing diverse OCR transcription errors and reduce the word error rate of 32.3% by more than 89%.
%R 10.1162/tacl_a_00379
%U https://aclanthology.org/2021.tacl-1.29
%U https://doi.org/10.1162/tacl_a_00379
%P 479-493

Markdown (Informal)

[Neural OCR Post-Hoc Correction of Historical Corpora](https://aclanthology.org/2021.tacl-1.29) (Lyu et al., TACL 2021)

• Neural OCR Post-Hoc Correction of Historical Corpora (Lyu et al., TACL 2021)

ACL

• Lijun Lyu, Maria Koutraki, Martin Krickl, and Besnik Fetahu. 2021. Neural OCR Post-Hoc Correction of Historical Corpora. Transactions of the Association for Computational Linguistics, 9:479–493.