Star Radiation Classification Using Six Categories of Stars with NASA Dataset Using Machine Learning Techniques
摘要
This study seeks the classification of stars into six classes according to stellar properties using these machine learning techniques. The features used for developing the dataset include temperature, luminosity, radius, magnitude, star color, spectral class, and star type from astronomical observations. The target is to classify the stars into different categories of Brown Dwarf, Red Dwarf, White Dwarf, Main Sequence, Supergiant, and Hypergiant. Some of the classification algorithms that were investigated include: AdaBoost, Decision Tree, and Random Forest. The prediction accuracy of each model, with Random Forest Classifier marking a high degree of precision for star classification showing values in the recall and F1 score, stood out better than the rest. This paper shows findings of relationships between stellar attributes and their classes that contribute to the field of astrophysics and machine learning.