Machine learning techniques for classifying dangerous asteroids

Seyed Matin Malakouti; Mohammad Bagher Menhaj; Amir Abolfazl Suratgar

doi:10.1016/j.mex.2023.102337

Machine learning techniques for classifying dangerous asteroids

MethodsX. 2023 Aug 19:11:102337. doi: 10.1016/j.mex.2023.102337. eCollection 2023 Dec.

Authors

Seyed Matin Malakouti¹, Mohammad Bagher Menhaj¹, Amir Abolfazl Suratgar¹

Affiliation

¹ Distributed and Intelligent Optimization Research Laboratory, Department of Electrical Engineering, Amirkabir University of Technology, Tehran, Iran.

Abstract

There is an infinite number of objects in outer space, and these objects and asteroids might be harmful. Hence, it is wise to know what is surrounding us and what can harm us amongst those. Therefore, in this article, with the hyperparameters tuning of Extra Tree, Random Forest, Light Gradient Boosting Machine, Gradient Boosting, and Ada Boost, the hazards of asteroids around the Earth were classified, and the results of ROC Curves for these algorithms were compared.•Reviewing the list of NASA-certified asteroids classified as the nearest Earth object•Investigating the risk of asteroids with the help of Extra Tree, Random Forest, Light Gradient Boosting Machine, Gradient Boosting, and Ada Boost•Comparing the performance of machine learning algorithms in the classification of high-risk asteroids.

Keywords: Asteroids; Hazards of asteroids; Machine learning; NASA.