R/qcluster-package.R

In qcluster: Clustering via Quadratic Scoring

#' qcluster: Clustering via Quadratic Scoring
#'
#' \code{qcluster} provides tools for tuning clustering models, methods, and
#' algorithms by quadratic scoring with resampling.
#'
#' The package combines three main components:
#' \itemize{
#'   \item method generators such as \code{\link{mset_user}},
#'   \code{\link{mset_gmix}}, \code{\link{mset_kmeans}},
#'   \code{\link{mset_pam}}, and \code{\link{mbind}};
#'   \item bootstrap quadratic-score estimation, ranking, and selection through
#'   \code{\link{bqs}}, \code{\link{bqs_rank}}, and
#'   \code{\link{bqs_select}};
#'   \item direct Gaussian model-based clustering and scoring via
#'   \code{\link{gmix}} and \code{\link{qscore}}.
#' }
#'
#' A typical workflow is:
#' \enumerate{
#'   \item define a collection of candidate clustering methods with
#'   \code{mset_*()} and optionally combine them with \code{\link{mbind}};
#'   \item estimate bootstrap quadratic scores with \code{\link{bqs}};
#'   \item rank candidate methods with \code{\link{bqs_rank}} and extract
#'   selected full-data refits with \code{\link{bqs_select}}.
#' }
#'
#' The package also provides plotting and printing methods for fitted
#' \code{mbcfit} and \code{bqs} objects, together with the sample data set
#' \code{\link{banknote}}.
#'
#' @references
#' Coraggio, Luca and Pietro Coretto (2023). Selecting the number of
#' clusters, clustering models, and algorithms. A unifying approach based
#' on the quadratic discriminant score.
#' \emph{Journal of Multivariate Analysis}, Vol. 196(105181), 1-20.
#' doi: \doi{10.1016/j.jmva.2023.105181}
#'
#' @useDynLib qcluster, .registration = TRUE, .fixes = "C_"
#' @import cluster
#' @import doParallel
#' @import foreach
#' @import grDevices
#' @import graphics
#' @import iterators
#' @import methods
#' @import parallel
#' @import stats
#'
#' @docType package
#' @name qcluster
#' @keywords package
"_PACKAGE"