Machine Learning based Classifier Models for Detection of Celestial Objects

Abstract

The classification of celestial objects such as stars, galaxies, and quasars is one of astronomy's most difficult and fundamental problems. Due to the technological advancement of telescopes and observatories, the classification of large volumes of data must be automated. Various machine learning techniques are currently employed for accurate classification. In this paper, a comparison of the efficacy of various classification algorithms is presented. XGBoost seems to be the most effective.