Application of Machine Learning Techniques To Young Stellar Object Classification




Crompvoets, Breanna

Journal Title

Journal ISSN

Volume Title



Among the first observations released to the public from the James Webb Space Telescope (JWST) was a section of the star-forming region NGC 3324 known colloquially as the “Cos- mic Cliffs.” We build a photometric catalog of the region and analyze these data using the Probabilistic Random Forest machine learning method. We find 496 YSOs out of 19 497 total objects within the field, 474 of which have not been found in previous works. Using the obtained probabilities of objects being YSOs, we employ a Monte Carlo approach to determine a local star formation rate of 1 × 10^–4 M⊙/yr, for the region. We also find that the surface density of YSOs in the Cosmic Cliffs is largely coincident with column densi- ties derived from Herschel data, up to a column density of 1.37 × 10^22 cm–2. The newly determined number and spatial distribution of YSOs in the Cosmic Cliffs demonstrate that JWST is far more capable of detecting YSOs in dusty regions than Spitzer.



Astronomy, Star formation, Machine learning, Random Forest, Probabilistic Random Forest