Skip to main content
All Reviews
AstronomyWorth Reading
intermediate

Machine learning classification of new asteroid families members

Carruba, V. et al. (2020)

Published
May 28, 2020
Journal
Monthly Notices of the Royal Astronomical Society · Vol. 496 · No. 1
DOI
10.1093/mnras/staa1463

At a GlanceAI

Machine learning can identify asteroid family members with up to 97% agreement with standard clustering methods.

SummaryAI

The rapid growth in known asteroids makes traditional family-identification methods harder to scale efficiently. This study tests nine machine learning classifiers to identify new asteroid family members from their orbital properties, using previously known family members as training data. Among the methods compared, extremely randomized trees performed best, recovering up to 97% of the members found by the standard hierarchical clustering method. The results suggest machine learning can help keep asteroid family classification practical as survey data continue to expand.

Method SnapshotAI

The study compares nine supervised machine learning classification algorithms using asteroid proper orbital elements.

BackgroundAI

Familiarity with asteroid families, orbital elements, and basic machine learning classification is helpful.

By comparing several supervised machine-learning classifiers for asteroid family identification, this study showed that data-driven methods can effectively complement traditional clustering approaches. The techniques remain relevant today, as they readily scale to the rapidly expanding asteroid databases produced by modern surveys.

VC