An all-sky support vector machine selection of WISE YSO candidates

G. Marton, L. Tóth, R. Paladini, M. Kun, S. Zahorecz, P. McGehee, C. Kiss

Research output: Contribution to journalArticle

30 Citations (Scopus)

Abstract

We explored the AllWISE catalogue of the Wide-field Infrared Survey Explorer (WISE) mission and identified Young Stellar Object (YSO) candidates. Reliable 2MASS and WISE photometric data combined with Planck dust opacity values were used to build our data set and to find the best classification scheme. A sophisticated statistical method, the support vector machine (SVM) is used to analyse the multidimensional data space and to remove source types identified as contaminants (extragalactic sources, main-sequence stars, evolved stars and sources related to the interstellar medium). Objects listed in the SIMBAD data base are used to identify the already known sources and to train our method. A new all-sky selection of 133 980 Class I/II YSO candidates is presented. The estimated contamination was found to be well below 1 per cent based on comparison with our SIMBAD training set. We also compare our results to that of existing methods and catalogues. The SVM selection process successfully identified >90 per cent of the Class I/II YSOs based on comparison with photometric and spectroscopic YSO catalogues. Our conclusion is that by using the SVM, our classification is able to identify more known YSOs of the training sample than other methods based on colour-colour and magnitude-colour selection. The distribution of the YSO candidates well correlates with that of the Planck Galactic Cold Clumps in the Taurus-Auriga-Perseus-California region.

Original languageEnglish
Pages (from-to)3479-3488
Number of pages10
JournalMonthly Notices of the Royal Astronomical Society
Volume458
Issue number4
DOIs
Publication statusPublished - Mar 17 2016

Fingerprint

Wide-field Infrared Survey Explorer
sky
catalogs
color
education
main sequence stars
data bases
clumps
opacity
train
contaminants
contamination
dust
stars
support vector machine
method
young
pollutant
comparison

Keywords

  • Infrared: General
  • Infrared: Stars
  • Methods: Data analysis
  • Methods: Statistical
  • Stars: Pre-main-sequence
  • Stars: Protostars

ASJC Scopus subject areas

  • Space and Planetary Science
  • Astronomy and Astrophysics

Cite this

An all-sky support vector machine selection of WISE YSO candidates. / Marton, G.; Tóth, L.; Paladini, R.; Kun, M.; Zahorecz, S.; McGehee, P.; Kiss, C.

In: Monthly Notices of the Royal Astronomical Society, Vol. 458, No. 4, 17.03.2016, p. 3479-3488.

Research output: Contribution to journalArticle

Marton, G. ; Tóth, L. ; Paladini, R. ; Kun, M. ; Zahorecz, S. ; McGehee, P. ; Kiss, C. / An all-sky support vector machine selection of WISE YSO candidates. In: Monthly Notices of the Royal Astronomical Society. 2016 ; Vol. 458, No. 4. pp. 3479-3488.
@article{042a31f09e614a499a442240281b2f11,
title = "An all-sky support vector machine selection of WISE YSO candidates",
abstract = "We explored the AllWISE catalogue of the Wide-field Infrared Survey Explorer (WISE) mission and identified Young Stellar Object (YSO) candidates. Reliable 2MASS and WISE photometric data combined with Planck dust opacity values were used to build our data set and to find the best classification scheme. A sophisticated statistical method, the support vector machine (SVM) is used to analyse the multidimensional data space and to remove source types identified as contaminants (extragalactic sources, main-sequence stars, evolved stars and sources related to the interstellar medium). Objects listed in the SIMBAD data base are used to identify the already known sources and to train our method. A new all-sky selection of 133 980 Class I/II YSO candidates is presented. The estimated contamination was found to be well below 1 per cent based on comparison with our SIMBAD training set. We also compare our results to that of existing methods and catalogues. The SVM selection process successfully identified >90 per cent of the Class I/II YSOs based on comparison with photometric and spectroscopic YSO catalogues. Our conclusion is that by using the SVM, our classification is able to identify more known YSOs of the training sample than other methods based on colour-colour and magnitude-colour selection. The distribution of the YSO candidates well correlates with that of the Planck Galactic Cold Clumps in the Taurus-Auriga-Perseus-California region.",
keywords = "Infrared: General, Infrared: Stars, Methods: Data analysis, Methods: Statistical, Stars: Pre-main-sequence, Stars: Protostars",
author = "G. Marton and L. T{\'o}th and R. Paladini and M. Kun and S. Zahorecz and P. McGehee and C. Kiss",
year = "2016",
month = "3",
day = "17",
doi = "10.1093/mnras/stw398",
language = "English",
volume = "458",
pages = "3479--3488",
journal = "Monthly Notices of the Royal Astronomical Society",
issn = "0035-8711",
publisher = "Oxford University Press",
number = "4",

}

TY - JOUR

T1 - An all-sky support vector machine selection of WISE YSO candidates

AU - Marton, G.

AU - Tóth, L.

AU - Paladini, R.

AU - Kun, M.

AU - Zahorecz, S.

AU - McGehee, P.

AU - Kiss, C.

PY - 2016/3/17

Y1 - 2016/3/17

N2 - We explored the AllWISE catalogue of the Wide-field Infrared Survey Explorer (WISE) mission and identified Young Stellar Object (YSO) candidates. Reliable 2MASS and WISE photometric data combined with Planck dust opacity values were used to build our data set and to find the best classification scheme. A sophisticated statistical method, the support vector machine (SVM) is used to analyse the multidimensional data space and to remove source types identified as contaminants (extragalactic sources, main-sequence stars, evolved stars and sources related to the interstellar medium). Objects listed in the SIMBAD data base are used to identify the already known sources and to train our method. A new all-sky selection of 133 980 Class I/II YSO candidates is presented. The estimated contamination was found to be well below 1 per cent based on comparison with our SIMBAD training set. We also compare our results to that of existing methods and catalogues. The SVM selection process successfully identified >90 per cent of the Class I/II YSOs based on comparison with photometric and spectroscopic YSO catalogues. Our conclusion is that by using the SVM, our classification is able to identify more known YSOs of the training sample than other methods based on colour-colour and magnitude-colour selection. The distribution of the YSO candidates well correlates with that of the Planck Galactic Cold Clumps in the Taurus-Auriga-Perseus-California region.

AB - We explored the AllWISE catalogue of the Wide-field Infrared Survey Explorer (WISE) mission and identified Young Stellar Object (YSO) candidates. Reliable 2MASS and WISE photometric data combined with Planck dust opacity values were used to build our data set and to find the best classification scheme. A sophisticated statistical method, the support vector machine (SVM) is used to analyse the multidimensional data space and to remove source types identified as contaminants (extragalactic sources, main-sequence stars, evolved stars and sources related to the interstellar medium). Objects listed in the SIMBAD data base are used to identify the already known sources and to train our method. A new all-sky selection of 133 980 Class I/II YSO candidates is presented. The estimated contamination was found to be well below 1 per cent based on comparison with our SIMBAD training set. We also compare our results to that of existing methods and catalogues. The SVM selection process successfully identified >90 per cent of the Class I/II YSOs based on comparison with photometric and spectroscopic YSO catalogues. Our conclusion is that by using the SVM, our classification is able to identify more known YSOs of the training sample than other methods based on colour-colour and magnitude-colour selection. The distribution of the YSO candidates well correlates with that of the Planck Galactic Cold Clumps in the Taurus-Auriga-Perseus-California region.

KW - Infrared: General

KW - Infrared: Stars

KW - Methods: Data analysis

KW - Methods: Statistical

KW - Stars: Pre-main-sequence

KW - Stars: Protostars

UR - http://www.scopus.com/inward/record.url?scp=84965018480&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84965018480&partnerID=8YFLogxK

U2 - 10.1093/mnras/stw398

DO - 10.1093/mnras/stw398

M3 - Article

AN - SCOPUS:84965018480

VL - 458

SP - 3479

EP - 3488

JO - Monthly Notices of the Royal Astronomical Society

JF - Monthly Notices of the Royal Astronomical Society

SN - 0035-8711

IS - 4

ER -