Eclipse Assessment Using Distributed Gradient Boosted Decision Tree-Specific Machine Learning Model

In the observable universe, we have mainly two types of eclipses on Earth: solar eclipse and lunar eclipse. Solar eclipse is defined as the alignment of Moon between Sun and Earth. Lunar eclipse is defined as the alignment of Earth between Sun and moon. A very popular method of predicting the eclipses is using Saros series which is approximately a cycle of 18 years and 11 days. A solar eclipse is estimated after 9 years and 5.5 days after the occurrence of lunar eclipse and vise versa. Eclipses impact the movement of organisms on earth. Also these are great astronomical events which is needed to be studied. In this paper, we have used machine learning models like XGBoost, gradient boosting, decision tree, and random forest classifier. This paper includes training of these algorithms and there prediction accuracy.