Scaled radial axes for interactive visual feature selection: A case study for analyzing chronic conditions.

•We propose a new radial axes method to help visual backward feature selection.•Experts can incorporate domain knowledge to analyze classes through LMNN, NCA, etc.•The method reduces clutter in visualization compared to other radial axes plots.•We conducted different experiments with several public data sets.•We present a case study using high dimensional data of chronic medical conditions.