Unsupervised classification of eclipsing binary light curves through k-medoids clustering

TL;DR

This paper introduces an unsupervised k-medoids clustering approach to classify eclipsing binary star light curves, providing a more objective and scientifically meaningful grouping than traditional methods.

Contribution

The study applies k-medoids clustering to classify variable stars based on light curves, revealing two main groups aligned with physical star properties, outperforming existing schemes.

Findings

Two optimal groups of eclipsing binaries identified

Clustering correlates with star brightness and mass

Method improves classification objectivity

Abstract

This paper proposes k-medoids clustering method to reveal the distinct groups of 1,318 variable stars in the Galaxy based on their light curves, where each light curve represents the graph of brightness of the star against time. To overcome the deficiencies of subjective traditional classification, we separate the stars more scientifically according to their geometrical configuration and show that our approach outperforms the existing classification schemes in astronomy. It results in two optimum groups of eclipsing binaries corresponding to bright, massive systems and fainter, less massive systems.