MORFITT

Data (Zenodo) | Publication (arXiv / HAL / ACL Anthology)

Yanis LABRAK, Richard DUFOUR, Mickaël ROUVIER

or

We introduce MORFITT, the first multi-label corpus for the classification of specialties in the medical field, in French. MORFITT is composed of 3,624 summaries of scientific articles from PubMed, annotated in 12 specialties. The article details the corpus, the experiments and the preliminary results obtained using a classifier based on the pre-trained language model CamemBERT.

For more details, please refer to our paper:

MORFITT: A multi-label topic classification for French Biomedical literature (arXiv / HAL / ACL Anthology)

Key Features

Documents distribution

Multi-label distribution

Number of labels per document distribution

Co-occurences distribution

If you use HuggingFace Transformers

from datasets import load_dataset
dataset = load_dataset("qanastek/MORFITT")
print(dataset)

or

from datasets import load_dataset
dataset_base = load_dataset(
    'csv',
    data_files={
        'train': f"./train.tsv",
        'validation': f"./dev.tsv",
        'test': f"./test.tsv",
    },
    delimiter="\t",
)

License and Citation

The code is under Apache-2.0 License.

The MORFITT dataset is licensed under Attribution-ShareAlike 4.0 International (CC BY-SA 4.0). If you find this project useful in your research, please cite the following papers:

Yanis Labrak, Mickaël Rouvier, Richard Dufour. MORFITT : A multi-label corpus of French scientific articles in the biomedical domain. 30e Conférence sur le Traitement Automatique des Langues Naturelles (TALN) Atelier sur l'Analyse et la Recherche de Textes Scientifiques, Florian Boudin, Jun 2023, Paris, France. ⟨hal-04125879⟩

or using the bibtex:

@inproceedings{labrak:hal-04125879,
  TITLE = {{MORFITT : A multi-label corpus of French scientific articles in the biomedical domain}},
  AUTHOR = {Labrak, Yanis and Rouvier, Micka{\"e}l and Dufour, Richard},
  URL = {https://hal.science/hal-04125879},
  BOOKTITLE = {{30e Conf{\'e}rence sur le Traitement Automatique des Langues Naturelles (TALN) Atelier sur l'Analyse et la Recherche de Textes Scientifiques}},
  ADDRESS = {Paris, France},
  ORGANIZATION = {{Florian Boudin}},
  YEAR = {2023},
  MONTH = Jun,
  KEYWORDS = {BERT ; RoBERTa ; Transformers ; Biomedical ; Clinical ; Topics ; multi-labels ; BERT ; RoBERTa ; Transformers ; Biom{\'e}dical ; Clinique ; Sp{\'e}cialit{\'e}s ; multi-labels},
  PDF = {https://hal.science/hal-04125879/file/_ARTS___TALN_RECITAL_2023__MORFITT__Multi_label_topic_classification_for_French_Biomedical_literature%20%285%29.pdf},
  HAL_ID = {hal-04125879},
  HAL_VERSION = {v1},
}