Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

T: Fachverband Teilchenphysik

T 54: Gammaastronomie II

T 54.4: Vortrag

Mittwoch, 21. März 2018, 17:20–17:35, Philo-HS7

Classification of unassociated 3FGL sources with Machine Learning — •Simone Mender, Kai Brügge, Maximilian Nöthe, and Kevin Schmidt — TU Dortmund, Lehrstuhl für Experimentelle Physik Vb, Otto-Hahn-Straße 4a, 44227 Dortmund

Active Galactic Nuclei (AGN) are astrophysical objects, whose emission range covers the entire electromagnetic spectrum. The AGN unification model includes numerous subclasses. The most powerful of them are the blazars, which are subdivided into BL Lac and Flat Spectrum Radio Quasars. To explore their phenomenology and their cosmological evolution it is interesting to look at average spectral energy distributions for the different classes.

The aim is to classify as many objects as possible so that they can be included in the calculation of average spectral energy distributions. To perform this classification Machine Learning can be used. In this talk, ongoing work based on the Fermi 3FGL catalog will be presented. It will be shown how unassociated sources and blazar candidates of uncertain type can be classified. For this purpose, methods of supervised and unsupervised learning are compared.