I am a data scientist in Safety Line company which specialized in aviation safety.
I am interested in machine learning, unsupervised and supervised algorithms, feature selection and functional data analysis. These methods are applied to the analysis of data coming from aircraft's Quick Access Recorders for predicting operational risks in aviation and for optimizing fuel consumption.
Supervised classification and regression of multidimensional functional data
Variable selection and variable importance
Patent (in french). Title: Procédé d’analyse d’un vol d’un aéronef pour évaluer l’exposition à un risque donné dudit vol. March 10th 2015. Patent number: 15/51975
B. Gregorutti, B. Michel and P. Saint-Pierre (2015). Grouped variable importance with random forests and application to multiple functional data analysis. Computational Statistics & Data Analysis, 90, p. 15-35 arXiv Preprint
B. Gregorutti, B. Michel and P. Saint-Pierre (2016). Correlation and variable importance in random forests. In press in Statistics and Computing. arXiv Preprint
B. Gregorutti (2016). RFgroove: Importance Measure and Selection for Groups of Variables with Random Forests. R package version 1.1. Download link