A new automated tool for the spectral classification of OB stars

Abstract

(abridged) We develop a tool for the automated spectral classification of OB stars according to their sub-types. We use the regular Random Forest (RF) algorithm, the Probabilistic RF (PRF), and we introduce the KDE-RF method which is a combination of the Kernel-Density Estimation and the RF algorithm. We train the algorithms on the Equivalent Width (EW) of characteristic absorption lines (features) measured in high-quality spectra from large Galactic (LAMOST,GOSSS) and extragalactic surveys (2dF,VFTS) with available spectral-types and luminosity classes. We find that the overall accuracy score is $\sim$70% with similar results across all approaches. We show that the full set of 17 spectral lines is needed to reach the maximum performance per spectral class. We apply our model in other observational data sets providing examples of potential application of our classifier on real science cases. We find that it performs well for both single massive stars and for the companion massive stars in Be X-ray binaries. In addition, we propose a reduced 10-features scheme that can be applied to large data sets with lower S/N. The similarity in the performances of our models indicates the robustness and the reliability of the RF algorithm when it is used for the spectral classification of early-type stars. The score of $\sim$70% is high if we consider (a) the complexity of such multi-class classification problems, (b) the intrinsic scatter of the EW distributions within the examined spectral classes, and (c) the diversity of the training set since we use data obtained from different surveys with different observing strategies. In addition, the approach presented in this work, is applicable to data of different quality and of different format (e.g.,absolute or normalized flux) while our classifier is agnostic to the Luminosity Class of a star and, as much as possible, metallicity independent.

arXiv commentAccepted for publication in the A&A journal. The code for the application of our models is available through the GitHub repository https://github.com/EliasKyritsis/SpectralClassifier

References in corpus (43)

• GW170817: Observation of Gravitational Waves from a Binary Neutron Star Inspiral
• The SIMBAD astronomical database
• Progenitors of core-collapse supernovae
• Pre-Supernova Evolution of Massive Single and Binary Stars
• Be/X-ray binaries
• The VLT-FLAMES Tarantula Survey VIII. Multiplicity properties of the O-type star population
• A Spectroscopic Study of the Ancient Milky Way: F- and G-Type Stars in the Third Data Release of the Sloan Digital Sky Survey
• The Supernova Channel of Super-AGB Stars
• The VLT-FLAMES Tarantula Survey I: Introduction and observational overview
• Probabilistic Random Forest: A machine learning algorithm for noisy datasets
• The Evolutionary Status of Be Stars: Results from a Photometric Study of Southern Open Clusters
• The Effect of Starburst Metallicity on Bright X-Ray Binary Formation Pathways
• A 2dF survey of the Small Magellanic Cloud
• Star formation history and X-ray binary populations: the case of the Small Magellanic Cloud
• Massive stars at low metallicity: Evolution and surface abundances of O dwarfs in the SMC
• Decision Tree Classifiers for Star/Galaxy Separation
• Star-formation history and X-ray binary populations: the case of the Large Magellanic Cloud
• The transition mass-loss rate: Calibrating the role of line-driven winds in massive star evolution
• Identifying Galaxy Mergers in Observations and Simulations with Deep Learning
• Application of Convolutional Neural Networks for Stellar Spectral Classification
• The VLT-FLAMES Tarantula Survey XVIII. Classifications and radial velocities of the B-type stars
• Identifying galaxies, quasars, and stars with machine learning: A new catalogue of classifications for 111 million SDSS sources without spectra
• The ARAUCARIA project: Grid-Based Quantitative Spectroscopic Study of Massive Blue Stars in NGC55
• A Catalogue of OB Stars from LAMOST Spectroscopic Survey
• Be stars and binaries in the field of the SMC open cluster NGC330 with VLT-FLAMES
• Rotating massive O stars with non-spherical 2D winds
• Astroinformatics of galaxies and quasars: a new general method for photometric redshifts estimation
• Sequentially Triggered Star Formation in OB Associations
• Stellar Spectra Classification and Feature evaluation Based on Random Forest
• The multiplicity of massive stars: a 2016 view
• Modeling X-ray binary evolution in normal galaxies: Insights from SINGS
• The Data Big Bang and the Expanding Digital Universe: High-Dimensional, Complex and Massive Data Sets in an Inflationary Epoch
• Physical parameters of IPHAS-selected classical Be stars. (I. Determination procedure and evaluation of the results.)
• Optical spectra of 5 new Be/X-ray Binaries in the Small Magellanic Cloud and the link of the supergiant B[e] star LHA 115-S 18 with an X-ray source
• SDSS-DR12 Bulk Stellar Spectral Classification: Artificial Neural Networks Approach
• Towards Real-time Classification of Astronomical Transients
• Uncertain classification of Variable Stars: handling observational GAPS and noise
• Evolutionary tracks of massive stars during formation
• Identifying New X-ray Binary Candidates in M31 using Random Forest Classification
• Distances to Galaxies from the Brightest Stars in the Universe
• Pulsation-driven mean zonal and meridional flows in rotating massive stars
• Hα imaging for BeXRBs in the Small Magellanic Cloud
• Census of blue stars in SDSS DR8

Cited by in corpus (6)

• Revisiting the mass of open clusters with \emph{Gaia} data
• Data mining techniques on astronomical spectra data. II : Classification Analysis
• A machine-learning photometric classifier for massive stars in nearby galaxies I. The method
• Dark Dust III: The high-quality single-cloud reddening curve sample. Scrutinizing extinction curves in the Milky Way
• Chemically peculiar stars on the pre-main sequence
• Astrophotometric search for massive stars in the Milky way. Confronting Random Forest predictions with available spectroscopy