APPLICATIONS OF MACHINE LEARNING (ML) IN THE FIELD OF PLANETARY SCIENCES AND GEOSCIENCES

The purpose of this study is to analyze the types of machine learning (ML) programs and algorithms that are commonly used in planetary sciences and geosciences. Fueled by increasing computer power and algorithmic advances, machine learning algorithms have become powerful tools for finding patterns in the innumerable amount of data collected in the fields of astrophysics, astronomy, earth sciences, planetary geology, and space sciences. The most commonly used ML algorithms - in the field of planetary sciences and geosciences - were determined by studying and analyzing critical information involving machine learning. Machine learning tools are receiving wider applications in solving problems in geosciences, particularly in remote sensing. Global dust sources and ground-level air-borne particulate matter are being researched effectively through ML learning algorithms. The paper provides detailed explanations and distinctions between types of machine learning algorithms. The paper compares and differentiates each algorithm’s current applications in these studies and the algorithm’s best practices. Current research not only goes into detail about the machine learning used in the fields of planetary sciences and geosciences but also provides an analysis of how algorithms can be best effectively utilized and aligned with a plausible explanation of the subject matter. In addition, the “Big Data”, as appropriately coined to address remote sensing, satellite, Lidar, and field data-based information, pose a challenging task to interpret and predict or establish models. Given its tremendous potential to handle huge datasets with multiple parameters, ML is apparently proving to be a viable mechanism to generate a meaningful trend for researchers.