README.md

---
license: cc-by-nc-4.0
language:
- fr
base_model:
- google-bert/bert-base-multilingual-cased
pipeline_tag: text-classification
widget:
- text: >-
    MAEATAE, (Géogr. anc.) anciens peuples de l'île de la grande Bretagne ; ils
    étoient auprès du mur qui coupoit l'île en deux parties.
datasets:
- GEODE/GeoEDdA-TopoRel
---



# bert-base-multilingual-cased-geography-entry-classification


<!-- Provide a quick summary of what the model is/does. -->

This model is designed to classify geographic encyclopedia articles into Place, Person, or Other.
It is a fine-tuned version of the bert-base-multilingual-cased model.
It has been trained on [GeoEDdA-TopoRel](https://huggingface.co/datasets/GEODE/GeoEDdA-TopoRel), a manually annotated subset of the French *Encyclopédie ou dictionnaire raisonné des sciences des arts et des métiers par une société de gens de lettres (1751-1772)* edited by Diderot and d'Alembert (provided by the [ARTFL Encyclopédie Project](https://artfl-project.uchicago.edu)).




## Model Description

<!-- Provide a longer summary of what this model is. -->

- **Authors:** Bin Yang, [Ludovic Moncla](https://ludovicmoncla.github.io), [Fabien Duchateau](https://perso.liris.cnrs.fr/fabien.duchateau/) and [Frédérique Laforest](https://perso.liris.cnrs.fr/flaforest/) in the framework of the [ECoDA](https://liris.cnrs.fr/projet-institutionnel/fil-2025-projet-ecoda) and [GEODE](https://geode-project.github.io) projects
- **Model type:** Text classification
- **Repository:** [https://gitlab.liris.cnrs.fr/ecoda/encyclopedia2geokg](https://gitlab.liris.cnrs.fr/ecoda/encyclopedia2geokg)
- **Language(s) (NLP):** French
- **License:** cc-by-nc-4.0


## Class labels


The tagset is as follows:
- **Place**: encyclopedia entry describing the name of a place (such as a city, a river, a country, etc.)
- **Person**: encyclopedia entry describing the name of a people or community
- **Other**: encyclopedia entry describing any other type of entity (such as abstract geographic concepts, cross-references to other entries, etc.)


## Dataset


The model was trained using the [GeoEDdA-TopoRel](https://huggingface.co/datasets/GEODE/GeoEDdA-TopoRel) dataset.
The dataset is splitted into train, validation and test sets which have the following distribution of entries among classes: 

|   | Train | Validation | Test|
|---|:---:|:---:|:---:|
| Place | 1,800 | 225 | 225|
| Person | 200 | 25 | 25 |
| Misc | 200 | 25 | 25 |


## Evaluation


* Overall weighted-average model performances


|   | Precision | Recall | F-score |
|---|:---:|:---:|:---:|
|    | 0.980   | 0.978   | 0.979 | 



* Model performances (Test set)

|   | Precision | Recall | F-score | Support |
|---|:---:|:---:|:---:|:---:|
| Place    |  0.99  |  0.98  |  0.99 | 225 |
| Person   |  1.00  |  0.96  |  0.98 | 25 |
| Other     |  0.83  |  0.96  |  0.89 | 25 |





## How to Get Started with the Model

Use the code below to get started with the model.


```python
import torch
from transformers import pipeline, AutoTokenizer, AutoModelForSequenceClassification
device = torch.device("mps" if torch.backends.mps.is_available() else ("cuda" if torch.cuda.is_available() else "cpu"))

tokenizer = AutoTokenizer.from_pretrained("GEODE/bert-base-multilingual-cased-geography-entry-classification")
model = AutoModelForSequenceClassification.from_pretrained("GEODE/bert-base-multilingual-cased-geography-entry-classification")

pipe = pipeline("text-classification", model=model, tokenizer=tokenizer,  truncation=True, device=device)

samples = [
    "* ALBI, (Géog.) ville de France, capitale de l'Albigeois, dans le haut Languedoc : elle est sur le Tarn. Long. 19. 49. lat. 43. 55. 44.",
    "MAEATAE, (Géogr. anc.) anciens peuples de l'île de la grande Bretagne ; ils étoient auprès du mur qui coupoit l'île en deux parties. Cambden ne doute point que ce soit le Nortumberland.",
    "APPONDURE, s. f. terme de riviere ; mot dont on se sert dans la composition d'un train ; c'est une portion  de perche employée pour fortifier le chantier lorsqu'il est trop menu."
]

for sample in samples:
    print(pipe(sample))

# Output
[{'label': 'Place', 'score': 0.9984742999076843}]
[{'label': 'Person', 'score': 0.9927592277526855}]
[{'label': 'Other', 'score': 0.9885557293891907}]

```


## Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical limitations. -->

This model was trained entirely on French encyclopaedic entries classified as Geography and will likely not perform well on text in other languages or other corpora. 



## Acknowledgement

The authors are grateful to the [ASLAN project](https://aslan.universite-lyon.fr) (ANR-10-LABX-0081) of the Université de Lyon, for its financial support within the French program "Investments for the Future" operated by the National Research Agency (ANR).
Data courtesy the [ARTFL Encyclopédie Project](https://artfl-project.uchicago.edu), University of Chicago.