extractionCatalogs : data extractor for 19th and 20th centuries art exhibition catalogs

This two step pipeline (eScriptorium + python script) aims to create TEI encoded art exhibition catalogs and CSV tables containing their data.

After an automatic/manual transcription with eScriptorium OCR (optical carachter recognition), the user will use the python script contained in this repository. It contains a Jupyter notebook file which serves as an interface for executing it.

You may find complete and detailed instructions in the Guide - extraction de catalogues.ipynb file :

inputs/outputs

eScriptorium

• input : iiif manifest or pdf, jpeg, png, etc.

• output : XML-ALTO files

python script :

• input : XML-ALTO files produced by eScriptorium

• output : XML-TEI and CSV files

The script output is an XML-TEI file which combines all the ALTO transcriptions and sticks to the ODD created by Caroline Corbières.

The /exemples_output directory contains examples of data produced by this pipeline.

Instructions

For complete instructions, use the Jupyter notebook integrated guide. This pipeline has a pedagogic vocation, so the guide is designed to introduce students with no prior technical knowledge to Digital Humanities. The python script can be executed directly from the notebook, so students may work without using the terminal.

1. Automatic/manual transcription of a digitized catalog with eScriptorium. Segmentation must be conformant to Segmonto Ontology.

2. Script execution with the following command :

python3 run.py input output title

• python3 run.py : python execution commands
• input : path to the directory containing ALTO transcribed catalog pages
• output : path to the desired repository for the output (TEI + CSV)
• title : catalog title (will be used for directorys and XML internal IDs)

Installation

• Clone the repository: git clone https://github.com/Juliettejns/extractionCatalogs
• Create virtual environment: virtualenv -p python3 env
• Run the virtual env: source env/bin/activate
• Install the requirements: pip install -r requirements.txt
• Stop the virtual env: source env/bin/deactivate

Credits

This repository was first developed as a prototype by Juliette Janes in 2021. The project was further augmented by Esteban Sánchez Oeconomo in 2022.

Both former students in Digital Humanities at Paris Science-Lettres University and interns of the Artl@s project worked under the direction of Béatrice Joyeux-Prunel and Simon Gabay.

Licence

The code is CC-BY.

Cite this repository

Juliette Janes, Esteban Sánchez Oeconomo, Simon Gabay, Béatrice Joyeux-Prunel, extractionCatalogs: Python data extractor for exhibition catalogs, 2022, Paris: ENS Paris https://github.com/Juliettejns/TEIcatalogs/

Contacts

If you have any questions or remarks, please contact juliette.janes@chartes.psl.eu, esteban.sanchez.oeconomo@chartes.psl.eu or simon.gabay@unige.ch.