Episciences.org TEI export of elpub:4607 - ElPub - ELectronic PUBlishing, 2018-06-20, Connecting the Knowledge Commons: From Projects to Sustainable InfrastructureCCSD - EpisciencesAttribution 4.0 International (CC BY 4.0)

Episciences.org API platform

Automatic Subject Indexing and Classification Using Text Recognition and Computer-Based Analysis of Tables of ContentsJanPokorny2018-06-20 21:22:572018-06-20 21:57:28contributorOpenEditionPresspress@openedition.orgCCSDelpub:4607http://elpub.episciences.org/4607elpub:4607 - ElPub - ELectronic PUBlishing, 2018-06-20, Connecting the Knowledge Commons: From Projects to Sustainable InfrastructureAttribution 4.0 International (CC BY 4.0)Automatic Subject Indexing and Classification Using Text Recognition and Computer-Based Analysis of Tables of ContentsJanPokornyhal-01816705ElPub - ELectronic PUBlishingConnecting the Knowledge Commons: From Projects to Sustainable InfrastructureLong Papers2018-06-20T21:57:28+02:0010.4000/proceedings.elpub.2018.19Englishmachine learning systemcomputer-generated keywordslibrary automatizationtext miningcomputer-generated subject headings[SHS.INFO]Humanities and Social Sciences/Library and information sciences

International audience

This paper will describe a method for machine-based creation of high quality subject indexing and classification for both electronic and print documents using tables of contents (ToCs). The technology described here is primarily focused on electronic and print documents for which, because of technical or licensing reasons, it is not possible to index full text. However, the technology would also be useful for full text documents, because it could significantly enhance the accuracy and relevance of subject description by analyzing the structure of ToCs.

ENKI, o.p.s.