Machine learning techniques to select Be star candidates

Abstract

Context. Optical and infrared variability surveys produce a large number of high quality light curves. Statistical pattern recognition methods have provided competitive solutions for variable star classification at a relatively low computational cost. In order to perform supervised classification, a set of features is proposed and used to train an automatic classification system. Quantities related to the magnitude density of the light curves and their Fourier coefficients have been chosen as features in previous studies. However, some of these features are not robust to the presence of outliers and the calculation of Fourier coefficients is computationally expensive for large data sets. Aims. We propose and evaluate the performance of a new robust set of features using supervised classifiers in order to look for new Be star candidates in the OGLE-IV Gaia south ecliptic pole field. Methods. We calculated the proposed set of features on six types of variable stars and also on a set of Be star candidates reported in the literature. We evaluated the performance of these features using classification trees and random forests along with the K-nearest neighbours, support vector machines, and gradient boosted trees methods. We tuned the classifiers with a 10-fold cross-validation and grid search. We then validated the performance of the best classifier on a set of OGLE-IV light curves and applied this to find new Be star candidates. Results. The random forest classifier outperformed the others. By using the random forest classifier and colours criteria we found 50 Be star candidates in the direction of the Gaia south ecliptic pole field, four of which have infrared colours that are consistent with Herbig Ae/Be stars. Conclusions. Supervised methods are very useful in order to obtain preliminary samples of variable stars extracted from large databases. As usual, the stars classified as Be stars candidates must be checked for the colours and spectroscopic characteristics expected for them.

Catalogue of new Herbig Ae/Be and classical Be stars

“…Solarz et al 2017;Ksoll et al 2018), or artificial neural networks (Snider et al 2001;Hampton et al 2017). However, similar performances are achieved with most of the algorithms and it is evident that the output is mainly dominated by the quality of the training data (Pérez-Ortiz et al 2017;Pashchenko et al 2018, or Marton et al 2019 in a similar problem of identifying YSOs). Therefore, we decided to use a shallow artificial neural network as it has the advantage of flexibility and non-linearity, being able to describe very complex and subtle relations.…”

Machine learning search for variable stars

“…Elorrieta et al (2016) used machine learning to identify RRab stars in the VVV survey data (Minniti et al 2010). Pérez-Ortiz et al (2017) propose a set of light curve features robust to individual outlier measurements and use them to compare multiple machine learning algorithms on classified OGLE-III light curves. Zinn et al (2017) consider an original set of features suitable for characterizing non-periodic and quasi-periodic light curves: parameters of the damped random walk and quasi-periodic oscillation models.…”

Searching for Be star candidates within large databases

“…We used as training sample the following OGLE-III variable stars: Cepheids, RR Lyrae, δ Scuti, long-period variables, type II Cepheids, eclipsing binaries, a set of OGLE-III BeSC reported in the literature, and samples of BeSC reported in previous works. The best classifier to select BeSC was random forests ( [4]). We used this classifier, and later color criteria, to search for BeSC within the OGLE-IV Gaia South Ecliptic Pole Field data.…”