Fully Three-Dimensional Radial Visualization

We develop methodology for three-dimensional (3D) radial visualization (RadViz) of multidimensional datasets. The classical two-dimensional (2D) RadViz visualizes multivariate data in the 2D plane by mapping every observation to a point inside unit circle. Our tool, RadViz3D, distributes anchor points uniformly on 3D sphere. show that this uniform distribution provides best with minimal artificial visual correlation uncorrelated variables. However, can be placed exactly equi-distant from each other only five Platonic solids, so we provide these settings, and approximately via Fibonacci grid cases. methodology, implemented R package $radviz3d$, makes fully possible is shown improve ability nonlinear technique more faithfully displaying simulated as well crabs, olive oils wine Additionally, because naturally suited compositional data, use RadViz3D illustrate (i) chemical composition Longquan celadon ceramics their Jingdezhen imitation over centuries, (ii) US regional SARS-Cov-2 variants' prevalence Covid-19 pandemic during summer 2021 surge Delta variant.