Page 1 of 1

UPSILoN: AUtomated Classification of Periodic Variable Stars using MachIne LearNing

Posted: Thu Dec 31, 2015 11:31 pm
by Ada Coda
UPSILoN: AUtomated Classification of Periodic Variable Stars using MachIne LearNing

Abstract: UPSILoN (AUtomated Classification of Periodic Variable Stars using MachIne LearNing) classifies periodic variable stars such as Delta Scuti stars, RR Lyraes, Cepheids, Type II Cepheids, eclipsing binaries, and long-period variables (i.e. superclasses), and their subclasses (e.g. RR Lyrae ab, c, d, and e types) using well-sampled light curves from any astronomical time-series surveys in optical bands regardless of their survey-specific characteristics such as color, magnitude, and sampling rate. UPSILoN consists of two parts, one which extracts variability features from a light curve, and another which classifies a light curve, and returns extracted features, a predicted class, and a class probability. In principle, UPSILoN can classify any light curves having arbitrary number of data points, but using light curves with more than ~80 data points provides the best classification quality.

Credit: Kim, Dae-Won; Bailer-Jones, Coryn A. L.

Site: https://github.com/dwkim78/upsilon
http://adsabs.harvard.edu/abs/2016A%26A...587A..18K

Bibcode: 2015ascl.soft12019K

Preferred citation method: http://adsabs.harvard.edu/abs/2016A%26A...587A..18K

ID: ascl:1512.019