Taxonomic classification of asteroids

We classified the asteroids with a combination of linear discriminant analysis and a nearest neighbor classifier. We achieve a classification accuracy of 92% for known S-class asteroids and an accuracy of 85% for Ch-class asteroids with a known set of 328 asteroids. Given the three classification parameters, tentative class designations for 1668 previously unclassified asteroids are provided in the Mahlke taxonomy.

We also show that the photometric slope values vary significantly within asteroid classes, with a standard deviation three to four times the mean slope uncertainties. In conclusion, we show that the combination of photometry and spectroscopy can be useful in the taxonomic classification of asteroids observed by Gaia. Further studies of the surface roughness at different scales could help clarify the potential of the photometric slope in classification efforts.