Massive young stellar objects in the Local Group spiral galaxy M 33 identified using machine learning

We present a supervised machine learning classification of stellar populations in the Local Group spiral galaxy M\,33. The Probabilistic Random Forest (PRF) methodology, previously applied to NGC\,6822, utilises both near and far-IR features. It classifies sources into nine target classes: young objects (YSOs), oxygen- carbon-rich asymptotic giant branch stars, red super-giant active galactic nuclei, blue stars (e.g. O-, B- A-type main sequence stars), Wolf-Rayet Galactic foreground stars. Across 100 runs PRF classified 162,746 with an average estimated accuracy $\sim$\,86\,per\,cent, based on confusion matrices. identified 4985 YSOs across disk M\,33, applying density-based clustering analysis identify 68 star forming regions (SFRs) primarily galaxy's arms. SFR counterparts known H\,{\sc ii} were recovered, $\sim$\,91\,per\,cent SFRs spatially coincident molecular clouds literature. Using photometric measurements, as well NGC\,6822 established evolutionary benchmark, we employed novel approach combining ratios [H$\alpha$]$/$[24$\mu$m] [250$\mu$m]$/$[500$\mu$m] estimate relative status all M\,33 SFRs. Masses for each YSO ranging from 6\,$-$\,27\,M$_\odot$. these masses, formation rates direct counts 0.63\,M$_\odot$\,yr$^{-1}$ M\,33's SFRs, 0.79\,$\pm$\,0.16\,M$_\odot$\,yr$^{-1}$ its centre 1.42\,$\pm$\,0.16\,M$_\odot$\,yr$^{-1}$ globally.