README.md

In mkampert/rCOSA: Clustering Objects on Subsets of Attributes (COSA)

rCOSA

rCOSA is an R package. The main output is a cluster happy dissimilarity matrix that can serve as input for proximity analysis methods.

Installation

These are the commands to install and load rCOSA:

install.packages('devtools');
devtools::install_github('mkampert/rCOSA')

Example

A detailed overview on how to use the rCOSA package is given in the open-source article "rCOSA: A Software Package for Clustering Objects on Subsets of Attributes" in the Journal of Classification (2017, Vol 34, issue 3, pp. 514 - 547).

A quick basic example of code on how to use the rCOSA package is:

library(rCOSA)
data(ApoE3) # ?ApoE3
cosa_rslts <- cosa2(ApoE3)
hierclust(cosa_rslts$D) 

Latest Developments

For the latest developments, check out "Improved Strategies for Distance Based Clustering of Objects on Subsets of Attributes in High-Dimensional Data"

mkampert/rCOSA documentation built on Dec. 23, 2019, 8:21 p.m.