Stacking Ensemble Model for Celestial Object Classification: Galaxies, Stars and Quasars

In the field of astronomy, it is essential to classify celestial objects like stars, galaxies, and quasars based on their spectral characteristics. This spectral data provides valuable information about various properties, such as the elements present, temperature, density, and magnetic field. To tackle this classification task, we investigate the application of different classification and ensemble algorithms. The proposed approach uses a variety of machine learning classifiers, including logistic regression, support vector machines, k-nearest neighbors, decision trees, random forests, and XGBoost. These classifiers are combined to create a stacking classifier, which is then evaluated on its accuracy, precision, recall, F1 score, and support. The Stacking classifier demonstrates the highest accuracy, reaching an impressive 99.99% on the training data. Train Logloss is 0.011. The precision, recall, and f1 score values (all 1.00) indicate a robust classification capability a cross all classes of celestial objects. This outstanding accuracy means that it effectively identifies almost all celestial objects in the training data-set. Consequently, the Stacking model serves as a highly dependable and precise tool for recognizing galaxies, stars, and quasars based on their spectral characteristics.