Keppler Red Giants Classification using a Machine learning approach

Modern satellite and telescopic observations of astounding quality are revolutionizing the constellations, stars, and celestial populations, along with identifying and characterizing exoplanets. Asteroseismology, studying stars via observing periodic oscillations in them, has seen some major changes. Asteroseismic Data Analysis provides a thorough technical overview of this field. The primary goal of this research is to lay out research information for the binary recognition of Kepler Red Giants. Popular Machine Learning (ML) classifiers and boosting techniques such as K-Nearest Neighbor (KNN), K-Means, Logistic Regression (LR), AdaBoost, Gradient Boosting, Extra Trees Classifier, Support Vector Machine (SVM), Naive Bayes, and Random Forest (RF) along with Principle Component Analysis (PCA) were applied to the extracted features from the given dataset and evaluated using various metrics. For Red Giants categorization, Gradient-Boosting surpassed all other machine learning techniques, with an accuracy score of 0.9618 and a precision of 0.9524. It was followed by the ensemble learning approach i.e. Extra Trees Classifier with an accuracy score of 0.9543.