Stage Master

Internship context:
Emerging pollutants are compounds that have not been classified but with a high potential
impact on health and environment, released in environment through different sources. Their
persistence and accumulation may represent an immediate hazard and long-term health outcome.
Fluorine appears in many emergent pollutants as a persistent organic pollutant (POP). One is PFOS,
known to be an endocrine disruptor.
The idea in the project is to develop efficient NMR and ML techniques for their detection and
identification.

Details about the internship tasks:
In this context, algorithms developments have begun and the proposed internship would consist
in the enrichment of a database containing NMR experiments on fluorinated molecules (virtually
adding artefacts, noise ...).
The aim would be to apply existing processing pipelines (based on Machine Learning for the
analysis and identification of present fluorinated molecules) with different hyperparameters to
realize statistical tests and evaluate the various ML algorithms against noise and artefacts.
Skills acquired during the internship are: understanding of the pipelines of data preparation,
dimensionality reduction, shuffling and scaling of datasets, as well as ML algorithm used.

Required skills:
- Python knowledge
- Data Management
- Programmation