Section: Bilateral Contracts and Grants with Industry

Arcelor-Mittal

Participants : Christophe Biernacki, Clément Thery.

Subject: Supervised and semi-supervised classification on large data bases mixing qualitative and quantitative variables. Arcelor Mittal faced some quality problems in the steel production which lead to supervised and semisupervised classification involving (1) a small number of individuals comparing to the numbers of variables, (2) heterogeneous variables, typically categorical and continous variables and (3) potentially highly correlated variables. A PhD CIFRE grant started on May 2011 on this topic and has finished on July 8th 2015. It has led also the the CorReg package, available on the CRAN (https://cran.r-project.org/web/packages/CorReg/index.html ) and referenced on the Inria BIL application