## File: bclust.Rd

package info (click to toggle)
r-cran-e1071 1.7-3-1
 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112 \name{bclust} \alias{bclust} \alias{hclust.bclust} \alias{plot.bclust} \alias{centers.bclust} \alias{clusters.bclust} \title{Bagged Clustering} \usage{ bclust(x, centers=2, iter.base=10, minsize=0, dist.method="euclidian", hclust.method="average", base.method="kmeans", base.centers=20, verbose=TRUE, final.kmeans=FALSE, docmdscale=FALSE, resample=TRUE, weights=NULL, maxcluster=base.centers, ...) hclust.bclust(object, x, centers, dist.method=object$dist.method, hclust.method=object$hclust.method, final.kmeans=FALSE, docmdscale = FALSE, maxcluster=object$maxcluster) \method{plot}{bclust}(x, maxcluster=x$maxcluster, main, ...) centers.bclust(object, k) clusters.bclust(object, k, x=NULL) } \arguments{ \item{x}{Matrix of inputs (or object of class \code{"bclust"} for plot).} \item{centers, k}{Number of clusters.} \item{iter.base}{Number of runs of the base cluster algorithm.} \item{minsize}{Minimum number of points in a base cluster.} \item{dist.method}{Distance method used for the hierarchical clustering, see \code{\link{dist}} for available distances.} \item{hclust.method}{Linkage method used for the hierarchical clustering, see \code{\link{hclust}} for available methods.} \item{base.method}{Partitioning cluster method used as base algorithm.} \item{base.centers}{Number of centers used in each repetition of the base method.} \item{verbose}{Output status messages.} \item{final.kmeans}{If \code{TRUE}, a final kmeans step is performed using the output of the bagged clustering as initialization.} \item{docmdscale}{Logical, if \code{TRUE} a \code{\link{cmdscale}} result is included in the return value.} \item{resample}{Logical, if \code{TRUE} the base method is run on bootstrap samples of \code{x}, else directly on \code{x}.} \item{weights}{Vector of length \code{nrow(x)}, weights for the resampling. By default all observations have equal weight.} \item{maxcluster}{Maximum number of clusters memberships are to be computed for.} \item{object}{Object of class \code{"bclust"}.} \item{main}{Main title of the plot.} \item{\dots}{Optional arguments top be passed to the base method in \code{bclust}, ignored in \code{plot}.} } \description{ Cluster the data in \code{x} using the bagged clustering algorithm. A partitioning cluster algorithm such as \code{\link{kmeans}} is run repeatedly on bootstrap samples from the original data. The resulting cluster centers are then combined using the hierarchical cluster algorithm \code{\link{hclust}}. } \details{ First, \code{iter.base} bootstrap samples of the original data in \code{x} are created by drawing with replacement. The base cluster method is run on each of these samples with \code{base.centers} centers. The \code{base.method} must be the name of a partitioning cluster function returning a list with the same components as the return value of \code{\link{kmeans}}. This results in a collection of \code{iter.base * base.centers} centers, which are subsequently clustered using the hierarchical method \code{\link{hclust}}. Base centers with less than \code{minsize} points in there respective partitions are removed before the hierarchical clustering. The resulting dendrogram is then cut to produce \code{centers} clusters. Hence, the name of the argument \code{centers} is a little bit misleading as the resulting clusters need not be convex, e.g., when single linkage is used. The name was chosen for compatibility with standard partitioning cluster methods such as \code{\link{kmeans}}. A new hierarchical clustering (e.g., using another \code{hclust.method}) re-using previous base runs can be performed by running \code{hclust.bclust} on the return value of \code{bclust}. } \value{ \code{bclust} and \code{hclust.bclust} return objects of class \code{"bclust"} including the components \item{hclust}{Return value of the hierarchical clustering of the collection of base centers (Object of class \code{"hclust"}).} \item{cluster}{Vector with indices of the clusters the inputs are assigned to.} \item{centers}{Matrix of centers of the final clusters. Only useful, if the hierarchical clustering method produces convex clusters.} \item{allcenters}{Matrix of all \code{iter.base * base.centers} centers found in the base runs.} } \author{Friedrich Leisch} \references{ Friedrich Leisch. Bagged clustering. Working Paper 51, SFB Adaptive Information Systems and Modeling in Economics and Management Science'', August 1999. \url{http://epub.wu.ac.at/1272/1/document.pdf}} \seealso{\code{\link{hclust}}, \code{\link{kmeans}}, \code{\link{boxplot.bclust}}} \keyword{multivariate} \keyword{cluster} \examples{ data(iris) bc1 <- bclust(iris[,1:4], 3, base.centers=5) plot(bc1) table(clusters.bclust(bc1, 3)) centers.bclust(bc1, 3) }