The SDL Component Suite is an industry leading collection of components supporting scientific and engineering computing. Please visit the SDL Web site for more information....


Home MathPack Math2 Procedures and Functions MeanDistanceKNN
See also: FindCenters, EstimateByKNN, kMeansClustering 

MeanDistanceKNN

Unit:SDL_math2
Class:none
Declaration:procedure MeanDistanceKNN (InMat: TMatrix; kn: integer; var DistVec: TVector);

The procedure MeanDistanceKNN calculates the mean distance of the kn nearest neighbors of each sample in the matrix InMat. The rows of the matrix InMat are seen as objects (samples), the columns are regarded as variables (measurements) which define the data space.

The results are stored in the vector DistVec. The user has to assure that the vector DistVec is large enough to accept the distances of all objects (DistVec has to have at least InMat.NrOfRows rows). If the vector DistVec is too small the result will be truncated accordingly.


Last Update: 2023-Feb-06