Near-Earth Object Risk Scoring Using Classical Machine Learning

Key Points

• The study aims to compare classical machine-learning classifiers to rule-based definitions for identifying potentially hazardous near-Earth objects.
• Utilized publicly available Near-Earth Object data from NASA CNEOS and JPL Small-Body Database
• Implemented classical machine-learning classifiers
• Focused on interpretability and reproducibility
• Assessed the methodology without making predictions about real-world hazards
• Classical machine-learning classifiers were evaluated against the standard rule-based approach
• Emphasized the importance of interpretability and reproducibility in analysis
• No impact prediction or real-world hazard assessment claims were made