The methodology implemented in this software package requires computing individual values of the efficient influence function (or just the influence function for a parametric model). Since these values are already computed / made available, a likely simple extension would be to use the matrix of EIF values to perform supervised clustering, as originally proposed in K.S. Pollard & M.J. van der Laan, "Supervised Distance Matrices", SAGMB, 2008. Heatmaps/plots based on this approach would likely provide a nice, user-friendly visualization.
The methodology implemented in this software package requires computing individual values of the efficient influence function (or just the influence function for a parametric model). Since these values are already computed / made available, a likely simple extension would be to use the matrix of EIF values to perform supervised clustering, as originally proposed in K.S. Pollard & M.J. van der Laan, "Supervised Distance Matrices", SAGMB, 2008. Heatmaps/plots based on this approach would likely provide a nice, user-friendly visualization.