Hazardous Asteroids prediction using machine learning

This study presents a machine learning-based approach for predicting potentially hazardous asteroids using data from NASA's Near-Earth Object (NEO) database. The objective is to develop an accurate and efficient classification model that can assist in early detection and risk assessment of asteroid threats. The dataset undergoes comprehensive preprocessing, including data cleaning, feature selection, and normalization to improve model performance. Multiple machine learning algorithms are implemented and evaluated, including Logistic Regression, Decision Tree, Random Forest, Gradient Boosting, and XGBoost. Among these, XGBoost demonstrates superior performance due to its ability to handle complex patterns and imbalanced data effectively. The model is evaluated using key performance metrics such as accuracy, precision, recall, F1-score, and ROC-AUC, ensuring robust and reliable predictions. The results indicate that the proposed approach achieves high classification accuracy and effectively identifies hazardous asteroids. This work contributes to the field of space safety and planetary defense by providing a scalable and data-driven solution for asteroid hazard prediction. The developed system can be further extended with real-time data integration and deployment for practical applications in monitoring near-Earth objects.