J/A+A/657/A62ivo://CDS.VizieR/J/A+A/657/A62doi:10.26093/cds/vizier.36570062CDSKyritsis E.Maravelias G.Zezas A.Bonfini P.Kovlakas K.Reig P.2022-03-17T14:50:02Z2022-01-11T08:57:49ZCDS support team

CDS, Observatoire de Strasbourg, 11 rue de l'Universite, F-67000 Strasbourg, France

cds-question@unistra.frearly-type-starsob-starsstellar-spectral-typesmorgan-keenan-classificationAs an increasing number of spectroscopic surveys become available, an automated approach to spectral classification becomes necessary. Due to the significance of the massive stars, it is of great importance to identify the phenomenological parameters of these stars (e.g., the spectral type), which can be used as proxies to their physical parameters (e.g., mass and temperature). In this work, we aim to use the random forest (RF) algorithm to develop a tool for the automated spectral classification of OB-type stars according to their sub-types. We used the regular RF algorithm, the probabilistic RF (PRF), which is an extension of RF that incorporates uncertainties, and we introduced the KDE-RF method which is a combination of the kernel-density estimation and the RF algorithm. We trained the algorithms on the equivalent width (EW) of characteristic absorption lines measured in high-quality spectra (Signal-to-Noise (S/N)>50) from large Galactic (LAMOST, GOSSS) and extragalactic surveys (2dF, VFTS) with available spectral types and luminosity classes. By following an adaptive binning approach, we grouped the labels of these data in 11 spectral classes within the O2-B9 range. We examined which of the characteristic spectral lines (features) are more important for the classification based on a number of feature selection methods, and we searched for the optimal hyperparameters of the classifiers to achieve the best performance. From the feature-screening process, we find that the full set of 17 spectral lines is needed to reach the maximum performance per spectral class. We find that the overall accuracy score is ~70%, with similar results across all approaches. We apply our model in other observational data sets providing examples of the potential application of our classifier to 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, especially for data of similar quality to the training sample. In addition, we propose a reduced ten-features scheme that can be applied to large data sets with lower S/N~20-50. 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 ~70% is high if we consider (a) the complexity of such multiclass classification problems (i.e., 11 classes), (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 products from different surveys in terms of quality (e.g., different resolution) and different formats (e.g., absolute or normalized flux), while our classifier is agnostic to the luminosity class of a star, and, as much as possible, it is metallicity independent.2022A&A...657A..62Khttps://cdsarc.cds.unistra.fr/viz-bin/cat/J/A+A/657/A62CatalogResearchIsServedByTAP VizieR generic serviceIsServedByConesearch servicehttps://cds.unistra.fr/vizier-org/licences_vizier.htmlhttps://vizier.cds.unistra.fr/viz-bin/VizieR-2?-source=J/A+A/657/A62https://vizier.iucaa.in/viz-bin/VizieR-2?-source=J/A+A/657/A62http://vizieridia.saao.ac.za/viz-bin/VizieR-2?-source=J/A+A/657/A62https://vizier.cds.unistra.fr/viz-bin/votable?-source=J/A+A/657/A62https://vizier.iucaa.in/viz-bin/votable?-source=J/A+A/657/A62http://vizieridia.saao.ac.za/viz-bin/votable?-source=J/A+A/657/A62GETtext/xml+votablehttps://tapvizier.cds.unistra.fr/TAPVizieR/tapCone search capability for table J/A+A/657/A62/table6 (Spectral type classification of stars in the IACOB survey)https://vizier.cds.unistra.fr/viz-bin/conesearch/J/A+A/657/A62/table6?https://vizier.iucaa.in/viz-bin/conesearch/J/A+A/657/A62/table6?http://vizieridia.saao.ac.za/viz-bin/conesearch/J/A+A/657/A62/table6?GETtext/xml+votable180.050000true2.623647857.16547560.0055555555555555566/63 161 326 455 461 463 472 474 496 649 859 867 1196 1198 1277 1293 1302 1323 1332 1385 1430 1539 1558 1704 1737 1764 1766 1770-1771 1789-1790 1826 1842 1946 2259 2359 2371 2409 2467 2593 2664 2737 2774-2775 2790 2806 2948 2970 2976 3000 3044 3051 3058 3060 3132 3136-3137 3144 3214 3220 3228 3242 3251 3256 3261 3307 3319 3430 3595 3621 3632 3637 3660 3704 3750 3830 4266 5506 6715 6891 6894 7767 7839 8547 9054 9680 9847 9936 10081 10731 10822 11599 12472 12651 12865 12927 12932 12943 12971 12991 13040 13155 13258 13292 13457 13557 13559 13572 13611 13682 13709 13733 13780 13910 13953 14008 14013 14077 14264 14313 14381 14431 14484 14522 14527 14545 14576 14581 14625 14632 14636 14641 14643 14653 14662 14669 14682 14706 14764 14777 14784 14810 14824 14843 14911 14945 14986 15368-15369 15417 15478 15504 15568 15604 15625 15631 15672 15686 15745 15747 15881 16275 16443 17194 17312 17322 18566 19245 19309 19336 19780 20012 20854 20860 21105 21309 21364 21401 21432 21434-21436 21438-21439 21467 21635 21647 21655 21677 21729 21879 22070 22102 22167 22209 22216 22249 22253 22258 22357 22400 22750 22769 22772 22787 22790 22805 22850 22854 22859 22862 22884 22888-22889 22893 22907 22920 22958 23005 23007 23122 23281 23358 23442 23483 23485 23637 23669 23673 23675 23692 23714 23717 23769 23776 23860 23977 23982 24029 24059 24132 24139 24179 24410 24480 24494 25016 26879 27322 27574 28827 28845 28851 28897 29097 29149 29213 29238 29253 29283 29327 29876 30088 30114 30289 30325 30476 30482 30643 30693 30723 31292 31812 32095 32604 32738 36018 36121 36651 39922 40502 40595 40599 40685 40807 42980 44310 44319 44461 44822 44892 48767 48884 49049 49109 49149https://cdsarc.cds.unistra.fr/viz-bin/moc/J/A+A/657/A62?format=asciiOpticaldefaultJ/A+A/657/A62/table6Spectral type classification of stars in the IACOB surveyrecnoRecord number assigned by the VizieR team. Should Not be used for identification.meta.recordintIDObject's ID in the IACOB databasemeta.id;meta.maincharIACOBSpTypePublished spectral typesrc.spTypecharRFSpTypePredicted spectral class based on the RF methodsrc.spTypecharProbProbability of predicted spectral classsrc.spTypefloatConfLevelConfidence level flag (see Sec. 5.1 of the paper)meta.codecharB0Probability for class B0stat.fit.goodnessfloatB1Probability for class B1stat.fit.goodnessfloatB2Probability for class B2stat.fit.goodnessfloatB3-B4Probability for class B3-B4stat.fit.goodnessfloatB5-B7Probability for class B5-B7stat.fit.goodnessfloatB8Probability for class B8stat.fit.goodnessfloatB9Probability for class B9stat.fit.goodnessfloatO2-O6Probability for class O2-O6stat.fit.goodnessfloatO7Probability for class O7stat.fit.goodnessfloatO8Probability for class O8stat.fit.goodnessfloatO9Probability for class O9stat.fit.goodnessfloatSimbadNameSimbad column added by the CDSmeta.idcharnullable_RARight Ascension (J2000) from SIMBAD (not part of the original data)degpos.eq.ra;meta.main_DEDeclination (J2000) from SIMBAD (not part of the original data)degpos.eq.dec;meta.main