2nd Metis Science Meeting

Applying Supervised Learning Methods for Metis Data Validation

Not scheduled

30m

Auditorium nazionale

Metis Science

Speaker

Burtovoi, Aleksandr (Istituto Nazionale di Astrofisica (INAF))

Description

We present the application of supervised learning methods to classify Metis VL and UV images during data validation activities. By examining Metis data collected over the first three years of the Nominal Phase of the Solar Orbiter mission, we labeled various features present in the images. This labeled dataset was then used to train a Support Vector Classification (SVC) model and to predict image categories. To evaluate the performance of the trained model, we compared the predicted labels with the actual ones and assessed the model’s accuracy using a confusion matrix.