Multidimensional scaling analysis of the solar system objects

摘要

In this paper, we analyse the attributes of some well-known objects in the solar system using clustering methods. Objects such as planets (with/without an estimate of Planet Nine), dwarf planets and some candidates for dwarf planets are considered. Different distances such as the Canberra, Jaccard, Lorentzian and Manhattan are adopted to calculate the relations between all objects based on their characteristics. The high dimensional information can thus be represented in a smaller number of dimensions via the maps generated by the hierarchical clustering and multidimensional scaling. The numerical measures and the computational tools allow the visualization of the solar system objects in a locus reflecting their characteristics. (C) 2019 Elsevier B.V. All rights reserved.