Random forest automated supervised classification of Hipparcos periodic variable stars

TL;DR

This paper evaluates the effectiveness of using random forest algorithms for automated classification of Hipparcos periodic variable stars into 26 types, achieving high accuracy with 90-100% correct classification rates.

Contribution

It introduces a supervised classification approach utilizing random forests and a multi-stage scheme, identifying key attributes for accurate star type classification.

Findings

Achieved 90-100% correct classification accuracy.

Identified key attributes like period, amplitude, and colour index for classification.

Provided online dataset of training and predicted star types.

Abstract

We present an evaluation of the performance of an automated classification of the Hipparcos periodic variable stars into 26 types. The sub-sample with the most reliable variability types available in the literature is used to train supervised algorithms to characterize the type dependencies on a number of attributes. The most useful attributes evaluated with the random forest methodology include, in decreasing order of importance, the period, the amplitude, the V-I colour index, the absolute magnitude, the residual around the folded light-curve model, the magnitude distribution skewness and the amplitude of the second harmonic of the Fourier series model relative to that of the fundamental frequency. Random forests and a multi-stage scheme involving Bayesian network and Gaussian mixture methods lead to statistically equivalent results. In standard 10-fold cross-validation experiments, the rate of correct classification is between 90 and 100%, depending on the variability type. The main mis-classification cases, up to a rate of about 10%, arise due to confusion between SPB and ACV blue variables and between eclipsing binaries, ellipsoidal variables and other variability types. Our training set and the predicted types for the other Hipparcos periodic stars are available online.