object.distances function

Calculate distances between object vectors in a SOM

This function calculates the distance between objects using the distance functions, weights and other attributes of a trained SOM. This function is used in the calculation of the U matrix in function plot.missSOM using the type = "dist.neighbours" argument.

object.distances(kohobj, type = c("data", "ximp", "codes"))

Arguments

• kohobj: An object of class missSOM.
• type: Whether to calculate distances between the data objects, or the codebook vectors.

Returns

An object of class dist, which can be directly fed into (e.g.) a hierarchical clustering.

Examples

data(wines)

## Data with no missing values 
set.seed(7)
sommap <- imputeSOM(scale(wines), grid = somgrid(6, 4, "hexagonal"))
obj.dists <- object.distances(sommap, type = "data")
code.dists <- object.distances(sommap, type = "codes")

## Data with missing values 
X <- scale(wines)
X[1:5, 1] <- NaN
sommap <- imputeSOM(X, grid = somgrid(6, 4, "hexagonal"))
obj.dists <- object.distances(sommap, type = "ximp")
code.dists <- object.distances(sommap, type = "codes")

See Also

unit.distances, imputeSOM