Classification of Stars using Random Forest Machine Learning Algorithm

Abstract

The project entails a dataset that consists of 250 stars that are to be classified into six different categories. This classification is based on six features: temperature, spectral class, Color, Absolute Magnitude, luminosity, and Radius. The crux of the project is to find the relation between these features in such a way so that a Machine Learning Model can differentiate them into various categories: Red Dwarfs, Brown Dwarfs, White Dwarfs, Main Sequence stars, SuperGiants and Hyper Giants. The Random Forest classification method was used to train the model, and quality metrics of the model were calculated to find the origin and cause behind the errors obtained. The classification method is simple and holds many advantages over its contenders. Since the project was built with python, the code was simplified. Basic statistics such as correlation matrix and harmonic mean were implemented using libraries such as matplotlib, enabling easy execution.