A machine learned classifier for RR Lyrae in the VVV survey

Felipe Elorrieta, Susana Eyheramendy, Andrés Jordán, István Dékány, Márcio Catelan, Rodolfo Angeloni, Javier Alonso-García, Rodrigo Contreras-Ramos, Felipe Gran, Gergely Hajdu, Néstor Espinoza, Roberto K. Saito, Dante Minniti

Resultado de la investigación: Article

15 Citas (Scopus)

Resumen

Variable stars of RR Lyrae type are a prime tool with which to obtain distances to old stellar populations in the Milky Way. One of the main aims of the Vista Variables in the Via Lactea (VVV) near-infrared survey is to use them to map the structure of the Galactic Bulge. Owing to the large number of expected sources, this requires an automated mechanism for selecting RR Lyrae, and particularly those of the more easily recognized type ab (i.e., fundamental-mode pulsators), from the 106-107 variables expected in the VVV survey area. In this work we describe a supervised machine-learned classifier constructed for assigning a score to a Ks-band VVV light curve that indicates its likelihood of being ab-type RR Lyrae. We describe the key steps in the construction of the classifier, which were the choice of features, training set, selection of aperture, and family of classifiers. We find that the AdaBoost family of classifiers give consistently the best performance for our problem, and obtain a classifier based on the AdaBoost algorithm that achieves a harmonic mean between false positives and false negatives of ≈7% for typical VVV light-curve sets. This performance is estimated using cross-validation and through the comparison to two independent datasets that were classified by human experts.

Idioma originalEnglish
Número de artículoA82
PublicaciónAstronomy and Astrophysics
Volumen595
DOI
EstadoPublished - 1 nov 2016

Huella dactilar

classifiers
near infrared
light curve
galactic bulge
variable stars
education
apertures
family
harmonics
comparison

ASJC Scopus subject areas

  • Astronomy and Astrophysics
  • Space and Planetary Science

Citar esto

Elorrieta, F., Eyheramendy, S., Jordán, A., Dékány, I., Catelan, M., Angeloni, R., ... Minniti, D. (2016). A machine learned classifier for RR Lyrae in the VVV survey. Astronomy and Astrophysics, 595, [A82]. https://doi.org/10.1051/0004-6361/201628700
Elorrieta, Felipe ; Eyheramendy, Susana ; Jordán, Andrés ; Dékány, István ; Catelan, Márcio ; Angeloni, Rodolfo ; Alonso-García, Javier ; Contreras-Ramos, Rodrigo ; Gran, Felipe ; Hajdu, Gergely ; Espinoza, Néstor ; Saito, Roberto K. ; Minniti, Dante. / A machine learned classifier for RR Lyrae in the VVV survey. En: Astronomy and Astrophysics. 2016 ; Vol. 595.
@article{95fd4f2ed87d47fd8d19496724b9684e,
title = "A machine learned classifier for RR Lyrae in the VVV survey",
abstract = "Variable stars of RR Lyrae type are a prime tool with which to obtain distances to old stellar populations in the Milky Way. One of the main aims of the Vista Variables in the Via Lactea (VVV) near-infrared survey is to use them to map the structure of the Galactic Bulge. Owing to the large number of expected sources, this requires an automated mechanism for selecting RR Lyrae, and particularly those of the more easily recognized type ab (i.e., fundamental-mode pulsators), from the 106-107 variables expected in the VVV survey area. In this work we describe a supervised machine-learned classifier constructed for assigning a score to a Ks-band VVV light curve that indicates its likelihood of being ab-type RR Lyrae. We describe the key steps in the construction of the classifier, which were the choice of features, training set, selection of aperture, and family of classifiers. We find that the AdaBoost family of classifiers give consistently the best performance for our problem, and obtain a classifier based on the AdaBoost algorithm that achieves a harmonic mean between false positives and false negatives of ≈7{\%} for typical VVV light-curve sets. This performance is estimated using cross-validation and through the comparison to two independent datasets that were classified by human experts.",
keywords = "Methods: data analysis, Methods: statistical, Stars: variables: RR Lyrae, Techniques: photometric",
author = "Felipe Elorrieta and Susana Eyheramendy and Andr{\'e}s Jord{\'a}n and Istv{\'a}n D{\'e}k{\'a}ny and M{\'a}rcio Catelan and Rodolfo Angeloni and Javier Alonso-Garc{\'i}a and Rodrigo Contreras-Ramos and Felipe Gran and Gergely Hajdu and N{\'e}stor Espinoza and Saito, {Roberto K.} and Dante Minniti",
year = "2016",
month = "11",
day = "1",
doi = "10.1051/0004-6361/201628700",
language = "English",
volume = "595",
journal = "Astronomy and Astrophysics",
issn = "0004-6361",
publisher = "EDP Sciences",

}

Elorrieta, F, Eyheramendy, S, Jordán, A, Dékány, I, Catelan, M, Angeloni, R, Alonso-García, J, Contreras-Ramos, R, Gran, F, Hajdu, G, Espinoza, N, Saito, RK & Minniti, D 2016, 'A machine learned classifier for RR Lyrae in the VVV survey', Astronomy and Astrophysics, vol. 595, A82. https://doi.org/10.1051/0004-6361/201628700

A machine learned classifier for RR Lyrae in the VVV survey. / Elorrieta, Felipe; Eyheramendy, Susana; Jordán, Andrés; Dékány, István; Catelan, Márcio; Angeloni, Rodolfo; Alonso-García, Javier; Contreras-Ramos, Rodrigo; Gran, Felipe; Hajdu, Gergely; Espinoza, Néstor; Saito, Roberto K.; Minniti, Dante.

En: Astronomy and Astrophysics, Vol. 595, A82, 01.11.2016.

Resultado de la investigación: Article

TY - JOUR

T1 - A machine learned classifier for RR Lyrae in the VVV survey

AU - Elorrieta, Felipe

AU - Eyheramendy, Susana

AU - Jordán, Andrés

AU - Dékány, István

AU - Catelan, Márcio

AU - Angeloni, Rodolfo

AU - Alonso-García, Javier

AU - Contreras-Ramos, Rodrigo

AU - Gran, Felipe

AU - Hajdu, Gergely

AU - Espinoza, Néstor

AU - Saito, Roberto K.

AU - Minniti, Dante

PY - 2016/11/1

Y1 - 2016/11/1

N2 - Variable stars of RR Lyrae type are a prime tool with which to obtain distances to old stellar populations in the Milky Way. One of the main aims of the Vista Variables in the Via Lactea (VVV) near-infrared survey is to use them to map the structure of the Galactic Bulge. Owing to the large number of expected sources, this requires an automated mechanism for selecting RR Lyrae, and particularly those of the more easily recognized type ab (i.e., fundamental-mode pulsators), from the 106-107 variables expected in the VVV survey area. In this work we describe a supervised machine-learned classifier constructed for assigning a score to a Ks-band VVV light curve that indicates its likelihood of being ab-type RR Lyrae. We describe the key steps in the construction of the classifier, which were the choice of features, training set, selection of aperture, and family of classifiers. We find that the AdaBoost family of classifiers give consistently the best performance for our problem, and obtain a classifier based on the AdaBoost algorithm that achieves a harmonic mean between false positives and false negatives of ≈7% for typical VVV light-curve sets. This performance is estimated using cross-validation and through the comparison to two independent datasets that were classified by human experts.

AB - Variable stars of RR Lyrae type are a prime tool with which to obtain distances to old stellar populations in the Milky Way. One of the main aims of the Vista Variables in the Via Lactea (VVV) near-infrared survey is to use them to map the structure of the Galactic Bulge. Owing to the large number of expected sources, this requires an automated mechanism for selecting RR Lyrae, and particularly those of the more easily recognized type ab (i.e., fundamental-mode pulsators), from the 106-107 variables expected in the VVV survey area. In this work we describe a supervised machine-learned classifier constructed for assigning a score to a Ks-band VVV light curve that indicates its likelihood of being ab-type RR Lyrae. We describe the key steps in the construction of the classifier, which were the choice of features, training set, selection of aperture, and family of classifiers. We find that the AdaBoost family of classifiers give consistently the best performance for our problem, and obtain a classifier based on the AdaBoost algorithm that achieves a harmonic mean between false positives and false negatives of ≈7% for typical VVV light-curve sets. This performance is estimated using cross-validation and through the comparison to two independent datasets that were classified by human experts.

KW - Methods: data analysis

KW - Methods: statistical

KW - Stars: variables: RR Lyrae

KW - Techniques: photometric

UR - http://www.scopus.com/inward/record.url?scp=84994669203&partnerID=8YFLogxK

U2 - 10.1051/0004-6361/201628700

DO - 10.1051/0004-6361/201628700

M3 - Article

AN - SCOPUS:84994669203

VL - 595

JO - Astronomy and Astrophysics

JF - Astronomy and Astrophysics

SN - 0004-6361

M1 - A82

ER -

Elorrieta F, Eyheramendy S, Jordán A, Dékány I, Catelan M, Angeloni R y otros. A machine learned classifier for RR Lyrae in the VVV survey. Astronomy and Astrophysics. 2016 nov 1;595. A82. https://doi.org/10.1051/0004-6361/201628700