bert-base-multilingual-cased-single-multiple-place-classification

This model is designed to classify geographic encyclopedia articles describing places. It is a fine-tuned version of the bert-base-multilingual-cased model. It has been trained on 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).

Model Description

• Developed by: Bin Yang, Ludovic Moncla, Fabien Duchateau and Frédérique Laforest
• Model type: Text classification
• Repository:
• Language(s) (NLP): French
• License: cc-by-nc-4.0

Class labels

The tagset is as follows:

• single: only one place is described
• multiple: several places are described (a single name with multiple locations)

Dataset

The model was trained using a set of 8658 entries classified as 'Place' (using this model: https://huggingface.co/GEODE/bert-base-multilingual-cased-geography-entry-classification) among entries classified as 'Geography' (using this model: https://huggingface.co/GEODE/bert-base-multilingual-cased-edda-domain-classification). The datasets have the following distribution of entries among datasets and classes:

	Train	Validation	Test
Single	5760	1235	1234
Multiple	300	64	65

Evaluation

• Overall macro-average model performances

Precision	Recall	F-score
0.92	0.92	0.92

• Overall weighted-average model performances

Precision	Recall	F-score
0.98	0.98	0.98

• Model performances (Test set)

	Precision	Recall	F-score	Support
Multiple	0.85	0.85	0.85	65
Single	0.99	0.99	0.99	1234

How to Get Started with the Model

Use the code below to get started with the model.

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-single-multiple-place-classification")
model = AutoModelForSequenceClassification.from_pretrained("GEODE/bert-base-multilingual-cased-single-multiple-place-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.",
    "PEGOE, (Géog. anc.) 1°. ville de l'Achaie, dans la Mégaride ; 2°. ville de l'Hellespont, selon Ortelius ; 3°. ville de l'île de Cypre ou de la Cyrénie, selon Etienne le géographe. "
]


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

Bias, Risks, and Limitations

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

Acknowledgement

The authors are grateful to the ASLAN project (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, University of Chicago.