CBL - Campus del Baix Llobregat

Projecte llegit

Títol: Random Forest identification of Gaia white dwarf-main sequence binary spectra

Estudiants que han llegit aquest projecte:

Director/a: TORRES GIL, SANTIAGO

Departament: FIS

Títol: Random Forest identification of Gaia white dwarf-main sequence binary spectra

Data inici oferta: 18-07-2022     Data finalització oferta: 18-03-2023


Estudis d'assignació del projecte:
    GR ENG SIST AEROESP
Tipus: Individual
 
Lloc de realització: EETAC
 
Segon director/a (UPC): REBASSA MANSERGAS, ALBERTO
 
Paraules clau:
deep learning, machine learning, space missions
 
Descripció del contingut i pla d'activitats:
Recent space missions such as the ESA Gaia satellite have
provided an unprecedented wealth of astronomical information for
billions of stars in our Galaxy. In particular, the third data
release of Gaia has provided spectra of nearly 200 million
sources. This huge quantity of data surpasses human capability to
be manually analyzed. Deep learning and machine learning
algorithms are therefore required. In the present project we will
apply a Random Forest algorithm for the spectral identification
of a particularly interesting type of binary star composed of a
white dwarf and a main-sequence star. This algorithm has been
successfully tested in a theoretical way in a previous TFG
project. We will now apply it to the real spectra recently
provided by Gaia, taking also into account all possible
observational characteristics such as contaminating objects,
blending effects, etc.
 
Overview (resum en anglès):

Recent space missions such as the ESA Gaia satellite have provided an unprecedented
wealth of astronomical information for billions of stars in our Galaxy. In particular, the
third data release of Gaia has provided spectra of nearly 200 million sources. This huge
quantity of data surpasses human capability to be manually analyzed. Deep learning and
machine learning algorithms are therefore required. In the present project we will apply a
Random Forest algorithm for the spectral identification of a particularly interesting type of
binary star composed of a white dwarf and a main-sequence star. This algorithm has been
successfully tested in a theoretical way in a previous Bachelor Thesis project. We will now
apply it to the real spectra recently provided by Gaia, taking also into account all possible
observational characteristics and for samples where visual identification reaches its limit.

© CBLTIC Campus del Baix Llobregat - UPC