README.md

---
license: lgpl
multilinguality:
- monolingual
language:
- fr
task_categories:
- other
task_ids:
- word-sense-disambiguation
dataset_info:
  features:
  - name: document_id
    dtype: string
  - name: sentence
    dtype: string
  - name: sentence_label
    dtype: string
  - name: sentence_first_label
    dtype: string
  - name: surface_forms
    sequence: string
  - name: labels
    sequence: string
  - name: first_labels
    sequence: string
  - name: word_id
    sequence: string
  - name: scores
    sequence: string
  - name: lemmas
    sequence: string
  - name: pos
    sequence: string
  splits:
  - name: SemCor
    num_bytes: 71632913
    num_examples: 37176
  - name: SemEval
    num_bytes: 749832
    num_examples: 306
  - name: WNGT
    num_bytes: 206691837
    num_examples: 117659
  download_size: 41831981
  dataset_size: 279074582
---

# Word Sense Disambiguation for FLUE 

## Dataset Description

- **Homepage:** 
- **Repository:** 
- **https://arxiv.org/pdf/1905.05677.pdf** 
- **Leaderboard:** 
- **loic.vial@univ-grenoble-alpes.fr** 

### Dataset Summary

This dataset is splitted in 3 sub-datasets: FrenchSemEval-Task12, French WNGT and an automatic translation of SemCor. 

### Supported Tasks and Leaderboards

Word Sense Disambiguation for French.  

### Language

French


### Licensing Information

```
GNU Lesser General Public License
```

### Citation Information

```bibtex
@inproceedings{vial-etal-2019-sense,
    title = "Sense Vocabulary Compression through the Semantic Knowledge of {W}ord{N}et for Neural Word Sense Disambiguation",
    author = {Vial, Lo{\"\i}c  and
      Lecouteux, Benjamin  and
      Schwab, Didier},
    booktitle = "Proceedings of the 10th Global Wordnet Conference",
    month = jul,
    year = "2019",
    address = "Wroclaw, Poland",
    publisher = "Global Wordnet Association",
    url = "https://aclanthology.org/2019.gwc-1.14",
    pages = "108--117",
    abstract = "In this article, we tackle the issue of the limited quantity of manually sense annotated corpora for the task of word sense disambiguation, by exploiting the semantic relationships between senses such as synonymy, hypernymy and hyponymy, in order to compress the sense vocabulary of Princeton WordNet, and thus reduce the number of different sense tags that must be observed to disambiguate all words of the lexical database. We propose two different methods that greatly reduce the size of neural WSD models, with the benefit of improving their coverage without additional training data, and without impacting their precision. In addition to our methods, we present a WSD system which relies on pre-trained BERT word vectors in order to achieve results that significantly outperforms the state of the art on all WSD evaluation tasks.",
}

```

### Contributions

* loic.vial@univ-grenoble-alpes.fr
* benjamin.lecouteux@univ-grenoble-alpes.fr
* didier.schwab@univ-grenoble-alpes.fr