19th-century French Land Registry : Information Extraction and Semantic Interpretation
Published May 23, 2025 | Version 0
Model Open

A fine-tuned NER model to structure taxpayers mentions extracted from the 19th-century french land registry

Authors/Creators

  • 1. ROR icon Laboratoire en Sciences et Technologies de l'Information Géographique pour la ville intelligente et les territoires durables
  • 2. ROR icon Institut national de l'information géographique et forestière
  • 3. EPITA

Description

This repository contains the weights of a CamemBERT model fine-tuned to perform named entities recognition on mentions of the taxpayers in the 19th century French Land Registry tables (initial registers). 

The dataset used to train and evaluate the model is available on Zenodo : 10.5281/zenodo.15423884. Annotation model is described in the dataset repository.

transformer Python library has been used to train the model.

The pre-trained CamemBERT model used to produce this fine-tuned version is available on Hugging Face : Jean-Baptiste/camembert-ner.

Files

NER-19lr-ir-94.zip

Files (1.0 GB)

Name Size Download all
md5:abea73bb294cbcb1b1e92c6439273922
1.0 GB Preview Download

Additional details

Software

Repository URL
https://github.com/solenn-tl/land_registry_tables_processing_TPDL_2025