DirectedDistanceMatrix[list, (options)] computes the symmetric or asymmetric matrix of distances or dissimilarity coefficients between the elements of list.
  • To use DirectedDistanceMatrix, you first need to load the Obtuse Package using Needs["Obtuse`"].
  • The elements of list can be numeric lists, matrices, or tensors, lists of Boolean elements, or strings. All data elements must have the same dimensions.
  • DirectedDistanceMatrix returns a symmetric matrix or an asymmetric matrix.
  • The method used to compute dissimilarities can be selected with the DistanceFunction option.
  • With the default setting DistanceFunction->Automatic, DirectedDistanceMatrix uses the square of the distance for numeric data, JaccardDissimilarity for Boolean data, and EditDistance for string data. The capabilities are a little beyond that. DirectedDistanceMatrix is capable of determining the distance between higher order tensors, and mixed numeric, string and Boolean data. Note, that since the square of the distance is used instead of the absolute value squared, the default setting is not suitable for points in the complex plane. This is a design choice - by using the square of the distance, the resulting matrix can be differentiated easily.
  • The setting for DistanceFunction can be any distance or dissimilarity function or a pure function f defining a distance between two values.
  • The following options can be given:
DistanceFunctionAutomaticdistance function to use in calculation of the distance matrix and the calculation of the distance from the interpolation point to the control points. Note, that the function here is expected to return the square of the distance