Application of Machine Learning Techniques for Heavy Rainfall Prediction using Satellite Data

This study presents a large-scale lunar dataset with the goal of utilizing machine learning techniques to forecast periods of intense rain. The dataset includes features from the lunar surface such as slope, height and depth of craters, types of boulders, patterns of distribution, and the presence of shadows. The study uses well-known machine learning techniques Random Forest, AdaBoost, Support Vector Machine (SVM), and k-Nearest Neighbors (KNN), with an emphasis on the Heavy Rain Prediction target variable, which indicates the chance of heavy rain events ($\mathrm{Yes} / \mathrm{No}$). In order to provide insights into efficient heavy rain event prediction on the moon’s surface, the algorithms are trained and assessed to identify patterns and linkages within the lunar dataset. The results further lunar exploration plans and provide useful recommendations for planning lunar missions and safety precautions against potentially unfavorable weather circumstances.