Skip to content

CompNet/FoppaInit

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

71 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

FoppaInit v1.0.3

Initialization of the FOPPA database

  • Copyright 2021-2024 Lucas Potin & Vincent Labatut

FoppaInit is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation. For source availability and license information see licence.txt


Description

These scripts create the FOPPA database v.1.1.3 from raw TED files. This database relies mainly on the award notices of public contracts related to French clients and suppliers from 2010 to 2020 in the Tenders Electronic Daily. It also proposes an enrichment of these data, thanks to the siretization of agents (i.e. the retrieval of their unique IDs, which is missing for most of them) as well as the cleaning and extraction of award criteria, and other processing.

The process conducted to build the FOPPA is quite long, though (around 1 week, depeding on the hardware), so the produced database is alternatively directly available on Zenodo. The detail of this processing are described in an article [P'23], and in further detail in a technical report [P'22].

This work was conducted in the framework of the DeCoMaP ANR project (Detection of corruption in public procurement markets -- ANR-19-CE38-0004). If you use this source code or the produced database, please cite bibliographical reference [P'23] (the article, not the report).

Organization

This repository is composed of the following elements:

  • requirements.txt : List of Python packages used in foppaInit.py.
  • foppaInit.py : Python script in order to create the FOPPA Database.
  • data : folder with the input files needed to create the database.

The script requires the TED data and leverages some additional sources:

Installation

You first need to install python and the required packages:

  1. Install the Python language: https://www.python.org
  2. Download this project from GitHub and unzip.
  3. Install CUDA: https://developer.nvidia.com/cuda-downloads
  4. Execute pip install -r requirements.txt to install some of the required packages (see also the Dependencies Section).

Package blazingsql requires a specific installation:

  1. Go to Webpage https://rapids.ai/start.html
  2. Configure your setup in this Webpage.
  3. Execute the generated command.

Use

In order to build the FOPPA database:

  1. Open the Python console.
  2. Run foppaInit.py.

The script is going to perform several tasks:

  1. Download all the necessary data (see Section Organization).
  2. Apply the processing described in [P'22].
  3. Export the resulting database under different forms (SQL dump, CSV sheets).

Dependencies

Tested with Python version 3.8.0, with the following packages:

Data

The produced database is directly available publicly online on Zenodo, under four different forms:

References

  • [P'23] L. Potin, V. Labatut, P. H. Morand & C. Largeron. FOPPA: An Open Database of French Public Procurement Award Notices From 2010–2020, Scientific Data, 2023, 10:303. DOI: 10.1038/s41597-023-02213-z ⟨hal-04101350⟩
  • [P'22] L. Potin, V. Labatut, R. Figueiredo, C. Largeron & P. H. Morand. FOPPA: A database of French Open Public Procurement Award notices, Technical Report, Avignon Université, 2022. ⟨hal-03796734⟩