Olivier Caelen

In September 2009, I got my PhD at the Machine Learning Group of the Université Libre de Bruxelles under the supervision of Professor Gianluca Bontempi. My research focused mainly on computational intelligence and in particular on the technics to speed up the models selection process. I am also interested in the multi armed bandit problem.

During my thesis, I worked on the three following projects:

• ARMURS project in collaboration with the IGEAT (Institut de Gestion de l'Environnement et de l'Aménagement du Territoire), the Royal Military Academy and the Laboratory of Image Synthesis and Analysis (ULB). Goal: building a system which uses remote sensing images and socio-economic data to assist topographical data producers in planning the update of their database more efficiently.
• OASIS+ project in collaboration with UCL (Université Catholique de Louvain), KUL (Katholieke Universiteit Leuven), the SMALS company, BCSS (Banque-Carrefour de la Sécurité Sociale) and SPF SS (Service Public Fédéral de la Sécurité Sociale). Goal: proposing a data-mining system to help to identify companies with a high probability of fraud.
• "Data mining prédictif en anesthésie intraveineuse informatisée" project in collaboration with the anesthesiology department of the Erasme Hospital and the Mexys SA. Company. Goal: developing a system for decision support in anesthesia.

Since November 2010, I work in the fraud detection team of Atos Worldline.
Follows a presentation given by my manager (S. Waterschoot): Fraud Risk Management in Practice.