Home   News   Projects   Skills   Software   Publications   Seminars   Links   Intranet  


Title: Detection and analysis of social fraud in OASIS Databases


Funding: Belgian Science Policy

Duration: 2007-2009 

Project page (Wiki): http://litpc45.ulb.ac.be/wikimlg

Workshop Page: http://litpc45.ulb.ac.be/OasisWorkshop

Project outline and goals:

The four federal department of social investigation (SPF Sécurité sociale, Office national de Sécurité sociale, SPF Emploi Travail et Concertation sociale, Office national de l'emploi) currently use the OASIS datawarehouse (Organisation Anti-fraude des Services d'Inspection Sociale). This datawarehouse centralizes administrative data from different federal services. These data are notably provided by the following sources :

  • DMFA (Déclaration multifonctionnelle - Multifunctionele aangifte)
  • Data about earnings and working time of workers (ONSS)
  • DIMONA (Déclaration immédiate - Onmiddelijke aangifte)
  • Data about duty starts and ends (ONSS)
  • Files of construction sites and subcontractors (ONSS)
  • List of employers (ONSS)
  • Accounts and collections files (ONEM)
  • Unemployed file
  • Quarterly statement of TVA (SPF Finances)
  • TVA Lists of customers and suppliers

We currently load into OASIS fiscal and social data regarding any employer and workers (anonymously) of building societies, for example. Besides, there is another file gathering results of controls. OASIS helps social investigators of the fraud squad to determine which company they should monitor. The system also provides social informations about companies. The main goal of OASIS is the ability to detect potential fraud as soon as possible. The selection of the companies is done using warnings that have been programed based on the experience of the investigators. Up to now, there is no datamining application that is either checking or valadating or upgrade these warnings. Furthermore, these data have been used by the investigating service only. The datawarehouse shows a strong potential that will lead, using datamining techniques, to a better understanding of what the social fraud is and to provide new informations to the persons in charge of this social domain.

This project has two main goals : upgrade OASIS by adding datamining tools and better understand the social fraud. Both goals are complementary. The dynamic modelisation provided by the datamining techniques will give a more precise of the (new) fraud mechanism. Then, the investigating services, the persons in charge and the researchers will better understand this social phenomenon in our country. Concretely, the project will produce an interface to the different communities (investigating services, public services, political services, scientists). Using a personalized interface, each one will refer and exploit data and results related to their needs.

MLG researchers involved: