---
language:
- fra
license:
- cc-by-nc-4.0
multilinguality:
- monolingual
source_datasets:
- original
task_categories:
- fill-mask
size_categories:
- 10M<n<100M
---

> [!NOTE]
> Dataset origin: https://github.com/CoFiF/Corpus

# Dataset Card for CoFiF    
    
    
## Dataset Description

- **Repository:** https://github.com/CoFiF/Corpus

- **Paper:** https://aclanthology.org/W19-5504/

- **Point of Contact:** Sina Ahmadi (sina.ahmadi@insight-centre.org) and Tobias Daudert (tobias.daudert@insight-centre.org)


### Dataset Summary

CoFiF is the first corpus comprising company reports in the French language. It contains over **188 million** tokens in **2655** reports, covering four types of documents:

- Reference documents (documents de référence) published annually, usually in the months following the end of the calendar year, and contain information regarding the financial situation and perspectives of a company
- aragnual report (résultats annuels) which summarises a company’s business and activities throughout the previous year
- smoemestrial (résultats semestriels): similar to annual reports in content but published every 6 months
- turimestrial reports (résultats trimestriels): similar to annual reports but published every 3 months

These documents are collected from the 60 largest French companies listed in France’s main stock indices [CAC40](https://en.wikipedia.org/wiki/CAC_40) and [CAC Next 20](https://en.wikipedia.org/wiki/CAC_Next_20). The corpus spans over 20 years, ranging from 1995 to 2018. 


### Supported Tasks and Leaderboards
Language modeling : can be used to train a French model (CamemBERT, FlauBERT, BARTHez, etc.) on data of financial types


### Languages
French



## Dataset Structure
Raw text



## Dataset Creation

### Curation Rationale
No French language datasets related to finance were available.

### Source Data
- Reference documents (documents de référence) published annually, usually in the months following the end of the calendar year, and contain information regarding the financial situation and perspectives of a company
- annual report (résultats annuels) which summarises a company’s business and activities throughout the previous year
- semestrial (résultats semestriels): similar to annual reports in content but published every 6 months
- trimestrial reports (résultats trimestriels): similar to annual reports but published every 3 months

These documents are collected from the 60 largest French companies listed in France’s main stock indices CAC40 and CAC Next 20. The corpus spans over 20 years, ranging from 1995 to 2018.


#### Initial Data Collection and Normalization
The authors have done a cleanup but have not detailed it in their publication or on their GitHub directory.

#### Who are the source language producers?
Public administrative reports written by humans.


### Annotations
No annotations.


## Additional Information

### Dataset Curators
This publication has emanated from research conducted with the financial support of Science Foundation Ireland (SFI) under Grant Number SFI/12/RC/2289, co-funded by the European Regional Development Fund.


### Licensing Information
The authors of the publication (Tobias Daudert and Sina Ahmadi) do not acknowledge any third party, so it appears that they were the only ones involved in the data collection.

This corpus is openly available for non-commercial use under the [Attribution-NonCommercial-ShareAlike 4.0 International](https://creativecommons.org/licenses/by-nc-sa/4.0/).


### Citation Information

If you're using CoFiF in your researches, please don't forget to cite [this paper](https://www.aclweb.org/anthology/papers/W/W19/W19-5504/):

~~~
@inproceedings{daudert-ahmadi-2019-cofif,
    title = "{C}o{F}i{F}: A Corpus of Financial Reports in {F}rench Language",
    author = "Daudert, Tobias  and Ahmadi, Sina",
    booktitle = "Proceedings of the First Workshop on Financial Technology and Natural Language Processing",
    month = "12 " # aug,
    year = "2019",
    address = "Macao, China",
    url = "https://www.aclweb.org/anthology/W19-5504",
    pages = "21--26",
}
~~~