Experimental Analysis of Stellar Classification by using Different Machine Learning Algorithms

In advanced Astrophysics, Machine Learning has progressively attained popularity with its astounding powerful potential to analyze large amounts of data and make accurate predictions. The categorization of stars is based on their spectral features, known as stellar classification. The star's electromagnetic energy is studied by dividing it using a diffraction or prism grating into a spectrum that has a rainbow of colours intermingled with spectral lines that are used for classification. The use of space-based imaging surveys necessitates a considerable deal of information on the spectral types of stars. This research presents the techniques for classifying stellar features into a Galaxy and a Star using various Machine Learning Algorithms on a dataset consisting of 100,000 observations. Techniques like K-Nearest Neighbours (KNN), Support Vector Classifier (SVC), Random Forest (RF), Logistic Regression (LR), Decision Tree (DT), and Naïve Bayes (NB) are used to train the model. Ultimately, a relative examination is done based on investigational outcomes from distinct models. This experiment demonstrates that the Support Vector Classifier (SVC) provides the highest accuracy, whereas the Decision tree (DT) provides the lowest. Finally, the potential for additional research in this area is highlighted.