Skip to contents

Create a hierarchical structure using a random hierarchical partition of the data.

Usage

hcRandomPairs(data, seed = NULL, ...)

Arguments

data

A numeric matrix or data frame of observations. If a matrix or data frame, rows correspond to observations and columns correspond to variables.

seed

Optional single value, interpreted as an integer, specifying the seed for random partition.

...

Catches unused arguments in indirect or list calls via do.call.

Value

A numeric two-column matrix in which the ith row gives the minimum index for observations in each of the two clusters merged at the

ith stage of a random agglomerative hierarchical clustering.

See also

Examples

data <- iris[,1:4]
randPairs <- hcRandomPairs(data)
str(randPairs)
#>  int [1:2, 1:149] 75 106 20 61 52 104 54 142 73 78 ...
#>  - attr(*, "initialPartition")= int [1:150] 1 2 3 4 5 6 7 8 9 10 ...
#>  - attr(*, "dimensions")= num [1:2] 150 2
# start model-based clustering from a random partition
mod <- Mclust(data, initialization = list(hcPairs = randPairs))
summary(mod)
#> ---------------------------------------------------- 
#> Gaussian finite mixture model fitted by EM algorithm 
#> ---------------------------------------------------- 
#> 
#> Mclust VVV (ellipsoidal, varying volume, shape, and orientation) model with 2
#> components: 
#> 
#>  log-likelihood   n df       BIC       ICL
#>       -214.3547 150 29 -574.0178 -574.0191
#> 
#> Clustering table:
#>   1   2 
#> 100  50