Josefina Barrera Morelli: Lectures on cheMOOmetrics: udderly accurate whey to predict

<p dir="ltr">My research is on the use of mid-infrared (MIR) spectral data to predict detailed milk composition, focussing on how the use of different techniques affect the results in regression models. For this purpose, I have evaluated various prepocessing techniques that are applied over the MIR spectra. Subsequently used the preprocessed data is combined with information on detailed composition to develop predictive models with algorithms (mostly machine learning) in my computer, to try to obtain the best accuracies possible. Whenever I have needed to explain my research to people the first things that come to mind are the “drawings” on the whiteboard of this cow and in more than one occasion I have wished to have a marker to do those myself to introduce my research to new people. So it felt fitting to illustrate this for the showcase.</p>