Automated analytical tools : API & artificial intelligence

A still in progress method is presented for the automated classification of marine phytoplankton groups observed with flow cytometry data. This method uses the cytometer files converted into JSON files thanks to a conversion tool built from an API meaning the constructor software isn't needed to have access to the cytometer files. The training set of the model was created with particles manually classified in a consensual way by different experts The model used is a Random Forest and it gives great results, an approach to the tuning of the model's parameters and the selection of the most important predictor variables, to optimise the process and the results, is also presented.