MétaCan
Menu
Back to cohort
Record W4306410260 · doi:10.35562/arabesques.3100

eScriptorium : une application libre pour la transcription automatique des manuscrits

2022· article· fr· W4306410260 on OpenAlexaff
Alix Chagué

Bibliographic record

VenueArabesques · 2022
Typearticle
Languagefr
FieldComputer Science
TopicMathematics, Computing, and Information Processing
Canadian institutionsUniversité de Montréal
Fundersnot available
KeywordsArt

Abstract

fetched live from OpenAlex

N°107 OCTOBRE -NOVEMBRE -DÉCEMBRE 2022 Ar ( abes ) ques 25 C ela fait longtemps que la transcription automatique des documents imprimés (OCR) et manuscrits (HTR) intéresse le monde de la recherche et celui des institutions patrimoniales.Le développement de processus s'appuyant sur l'intelligence artificielle et l'augmentation des capacités de calcul ont récemment ouvert de nouvelles perspectives.Dès le début des années 2000, des campagnes d'OCR ont été mises en place pour traiter les imprimés.Pour les manuscrits en revanche, ce n'est qu'à partir du milieu des années 2010 que les choses ont commencé à changer avec l'apparition de logiciels disponibles en ligne sur abonnement comme Transkribus ou en open source comme eScriptorium.C'est le groupe de recherche SCRIPTA PSL 1 qui développe, depuis 2019, l'application eScriptorium dont la vocation principale était de doter le logiciel Kraken 2 d'une interface graphique facilitant son utilisation.Kraken est un logiciel de transcription automatique développé en open source en 2015 par Benjamin Kiessling et conçu initialement pour proposer une meilleure prise en charge des textes non latins, en particulier arabes.Aujourd'hui, le groupe bénéficie des contributions d'autres infrastructures ou projets de recherche qui ont adopté l'application.Ce fut le cas du projet LectAuRep (Inria/Archives nationales) jusqu'en 2022 ou encore du groupe OpenITI (université du Maryland).

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.003
metaresearch head score (Gemma)0.014
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Not applicable · Consensus signal: none
GenreCandidate signal: Software · Consensus signal: none
Teacher disagreement score0.126
Threshold uncertainty score0.421

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0030.014
Meta-epidemiology (narrow)0.0040.002
Meta-epidemiology (broad)0.0020.002
Bibliometrics0.0050.002
Science and technology studies0.0010.001
Scholarly communication0.0050.004
Open science0.0020.004
Research integrity0.0030.002
Insufficient payload (model declined to judge)0.1260.103

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.026
GPT teacher head0.252
Teacher spread0.226 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designNot applicable
Domainnot available
GenreSoftware

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations1
Published2022
Admission routes1
Has abstractyes

Explore more

Same venueArabesquesSame topicMathematics, Computing, and Information ProcessingFrench-language works237,207