Estimates of the atmospheric parameters of M-type stars: A machine-learning perspective

  1. Sarro, L.M. 1
  2. Ordieres-Meré, J. 2
  3. Bello-García, A. 3
  4. González-Marcos, A. 4
  5. Solano, E. 5
  1. 1 Universidad Nacional de Educación a Distancia
    info

    Universidad Nacional de Educación a Distancia

    Madrid, España

    ROR https://ror.org/02msb5n36

  2. 2 Universidad Politécnica de Madrid
    info

    Universidad Politécnica de Madrid

    Madrid, España

    ROR https://ror.org/03n6nwv02

  3. 3 Universidad de Oviedo
    info

    Universidad de Oviedo

    Oviedo, España

    ROR https://ror.org/006gksa02

  4. 4 Universidad de La Rioja
    info

    Universidad de La Rioja

    Logroño, España

    ROR https://ror.org/0553yr311

  5. 5 Centro de Astrobiología
    info

    Centro de Astrobiología

    Madrid, España

    ROR https://ror.org/038szmr31

Revista:
Monthly Notices of the Royal Astronomical Society

ISSN: 0035-8711

Año de publicación: 2018

Volumen: 476

Número: 1

Páginas: 1120-1139

Tipo: Artículo

beta Ver similares en nube de resultados
DOI: 10.1093/MNRAS/STY165 SCOPUS: 2-s2.0-85047982132 GOOGLE SCHOLAR

Otras publicaciones en: Monthly Notices of the Royal Astronomical Society

Resumen

Estimating the atmospheric parameters of M-type stars has been a difficult task due to the lack of simple diagnostics in the stellar spectra. We aim at uncovering good sets of predictive features of stellar atmospheric parameters (Teff, log(g), [M/H]) in spectra of M-type stars. We define two types of potential features (equivalent widths and integrated flux ratios) able to explain the atmospheric physical parameters. We search the space of feature sets using a genetic algorithm that evaluates solutions by their prediction performance in the framework of the BT-Settl library of stellar spectra. Thereafter, we construct eight regression models using different machine-learning techniques and compare their performances with those obtained using the classical χ2 approach and independent component analysis (ICA) coefficients. Finally, we validate the various alternatives using two sets of real spectra from the NASA Infrared Telescope Facility (IRTF) and Dwarf Archives collections. We find that the crossvalidation errors are poor measures of the performance of regression models in the context of physical parameter prediction in M-type stars. For R ~ 2000 spectra with signal-to-noise ratios typical of the IRTF and Dwarf Archives, feature selection with genetic algorithms or alternative techniques produces only marginal advantages with respect to representation spaces that are unconstrained in wavelength (full spectrum or ICA). We make available the atmospheric parameters for the two collections of observed spectra as online material. © 2018 The Author(s).