• Fermi
• gamma-ray sources
• machine learning

The study of astrophysical sources of gamma rays can be very useful for analyzing the behavior of matter in extreme conditions, which are difficult to reproduce in the laboratory.
The Fermi telescope is currently the most sensitive gamma-ray telescope in orbit. Its technological improvements have led to a large increase in the number of detected sources.
However, in many cases these sources are still unidentified, i.e. we do not know what its nature is.
The identification process requires long and expensive observation campaigns, which can however be accelerated using new data analysis techniques, such as Machine Learning.
In this thesis we therefore analyze the potential and limits of various Machine Learning techniques for the classification of the unidentified sources of the Fermi telescope.