combineClusterings: An ensemble method for combining multiple clustering outcomes...

In rwoldford/trec: Tree Reduced Ensemble Clustering

Description

An ensemble method for combining multiple clustering outcomes based on monotonic graph families of Zhou and Oldford

Usage

combineClusterings(clustering1, clustering2, ..., labels = NULL,
  weights = NULL, pruneNumber = 1)

Arguments

clustering1	result of some clustering, for example output from hclust(). A clustering can also be an n by m matrix, where n is the number of data points and m is the number of levels in the clustering hierarchy.
clustering2	result of a second clustering, to be combined with the first.
...	results of other clustering methods, to be combined with the first two.
labels	labels of data points in clustering results
pruneNumber	set number for pruning trivial components

Value

a clusterTree object, which is the final clustering result from combining all input clustering results

Examples

data <- rbind(matrix(rnorm(100, mean = 10, sd = 2), nrow = 50),
              matrix(rnorm(100, mean = 0, sd = 1), nrow = 50),
              matrix(rnorm(100, mean = -10, sd = 3), nrow = 50)
              )
clustering1 <- stats::hclust(dist(data),method='single')
clustering2 <- kmeans(data,centers=3)
clustering3 <- dbscan::dbscan(data,eps=.8)
res <- combineClusterings(clustering1,clustering2,clustering3)

rwoldford/trec documentation built on May 15, 2019, 6:29 p.m.