I am a data scientist with a strong interest in feature selection. I focus on the developments of new techniques for mining large high-dimensional datasets. My current work concerns:
- a new index of clustering for structure detection and quantification.
- a new estimator of the intrinsic dimension of data.
- novel feature selection techniques based on data intrinsic dimension estimation.
- the development of the R package IDmining (available from the CRAN repository).
- a book about the machine learning of environmental data.
- deep learning for remote sensing scene classification (with the developments of open-source Python codes).
I got a master’s degree in environmental geosciences from the University of Lausanne in 2013 (with honors). My master’s thesis focused on the use of geostatistical simulations for the analysis and modelling of monitoring network data. Then I started a PhD research project under the supervision of Prof. Mikhail Kanevski, and I joined the doctoral program in statistics and applied probabilities. I am now a PhD candidate.