Skip to Content

French Office for the Protection of Refugees and Stateless Persons - OFPRA

With Arkhênum

France ✧ 2024

Project


Indexing the archives of Spanish Civil War refugees for better accessibility and preservation, as part of promoting democratic memory in Europe.


Corpus


The Spanish file contains 185,302 index cards in 121 drawers, listing Spanish refugees between 1945 and 1981.

350, 000 fiches (recto-verso), 
700 000 images in total 

Processing 


-> Automatic Text Recognition 

-> Information Extraction

Processing workflow


  • On-site digitization of color documents (by Arkhênum) 
  • Training the models via an annotation campaign on Callico
  • Naming, organizing scanned files and inserting metadata
  • Manual indexing or indexing by automatic processing
  • Matching the results obtained with OFPRA data
  • Export to CSV file and integration into the OFPRA database