Protostellar classification using supervised machine learning algorithms

Abstract

Classification of young stellar objects (YSOs) into different evolutionary stages helps us to understand the formation process of new stars and planetary systems. Such classification has traditionally been based on spectral energy distribution (SED) analysis. An alternative approach is provided by supervised machine learning algorithms, which can be trained to classify large samples of YSOs much faster than via SED analysis. We attempt to classify a sample of Orion YSOs (the parent sample size is 330) into different classes, where each source has already been classified using multiwavelength SED analysis. We used eight different learning algorithms to classify the target YSOs, namely a decision tree, random forest, gradient boosting machine (GBM), logistic regression, naïve Bayes classifier, k-nearest neighbour classifier, support vector machine, and neural network. The classifiers were trained and tested by using a 10-fold cross-validation procedure. As the learning features, we employed ten different continuum flux densities spanning from the near-infrared to submillimetre wavebands (λ = 3.6 − 870 µm). With a classification accuracy of 82% (with respect to the SED-based classes), a GBM algorithm was found to exhibit the best performance. The lowest accuracy of 47% was obtained with a naïve Bayes classifier. Our analysis suggests that the inclusion of the 3.6 µm and 24 µm flux densities is useful to maximise the YSO classification accuracy. Although machine learning has the potential to provide a rapid and fairly reliable way to classify YSOs, an SED analysis is still needed to derive the physical properties of the sources (e.g. dust temperature and mass), and to create the labelled training data. The machine learning classification accuracies can be improved with respect O. Miettinen 1 Digia Plc/Avarea Oy, Rautatieläisenkatu 6, FI-00520 Helsinki, Finland to the present results by using larger data sets, more detailed missing value imputation, and advanced ensemble methods (e.g. extreme gradient boosting). Overall, the application of machine learning is expected to be very useful in the era of big astronomical data, for example to quickly assemble interesting target source samples for follow-up studies.

Complementary dataset of the paper "The viewing angle in AGN SED models, a data-driven analysis"

“…Based on the median estimated errors, our results are better constrained than the ones from Vika et al (2017). 2016; Miettinen 2018;Jayasinghe et al 2018;Bluck et al 2020;Baqui et al 2021). Nowadays, these methods are helping to classify astrophysical objects not only from reduced fluxes, but also from astronomical imaging surveys, where Type-1 AGN are separated from normal galaxies (Golob et al 2021).…”

The viewing angle in AGN SED models: a data-driven analysis

“…These classification tasks have covered the full range of galactic and extragalactic sources (e.g. Tamayo et al 2016 ;Jayasinghe et al 2018 ;Miettinen 2018 ;Bluck et al 2020 ;Baqui et al 2021 ). Nowadays, these methods are helping to classify astrophysical objects not only from reduced fluxes, but also from astronomical imaging surv e ys, where Type-1 AGN are separated from normal galaxies (Golob et al 2021 ).…”

Surveying the reach and maturity of machine learning and artificial intelligence in astronomy

“…Spectroscopy provides information on the atomic and molecular composition, from which other physical properties (temperature, density, metallicity, etc.) can be inferred (e.g., Li et al, ; Márquez‐Neila, Fisher, Sznitman, & Heng, ; Miettinen, ; Ucci, Ferrara, Pallottini, & Gallerani, ).…”

“…Insight into candidate and confirmed extrasolar planets is also being achieved with ML and AI, such as through the determination of a “habitability score” for extrasolar planets (Saha et al, ) and improved model fitting of atmospheric composition (Márquez‐Neila et al, ). Stars and stellar products . Two key activities in stellar astronomy are spectral classification (e.g., Garcia‐Dias, Allende Prieto, Sánchez Almeida, & Ordovás‐Pascual, ; Wang, Guo, & Luo, with k ‐means clustering; Kong et al, ; classification of young stellar objects with eight different methods by Miettinen, ) and photometric classification (e.g., Ksoll et al, ; Zhang et al, with SVM, RF, and Fast Boxes). Many new examples of specific stellar classes have been discovered, such as Wolf‐Rayet stars (Morello et al, ), blue horizontal branch stars (Wan et al, ), hot sub dwarf stars (Bu et al, ), and rare hypervelocity stars (Marchetti et al, ).…”