CompareClustering: Compare Clusterings
In laderast/Consense: Consense: a package for evaluating multiple clusterings

Description Usage Arguments Value Author(s) See Also Examples

Main framework for comparing clustering functions.

1	CompareClustering(data, verbose = TRUE, auto = FALSE, kinput)

`data`	A matrix of values to cluster by row. To cluster by column, simply transpose the matrix. However, note that the row names must be unique, as the framework uses the row names in calculating the between-method metrics.
`verbose`	If TRUE, show progress of execution. If not, run without showing progress of execution.
`auto`	If FALSE, run getClusterInfo to query user as to parameters. If TRUE, use kinput as number of clusters for methodlistAuto
`kinput`	For automated clustering, a value for k (number of clusters) to be input to the clustering methods. If left blank or has a character value, the default mode is to allow users to hand cut their clusters using `inspectTree`.

`report`	A list containing the metrics. report$within contains a matrix of all the within-method metrics calculated. report$between contains the two matrices comparing all methods: between$jaccardresult contains the Jaccard Indices, and between$variationresult contains all Variation of information metrics.
`params`	A list of lists containing all parameters input to the clustering methods. Note that many of these parameters are unique to the clustering method. params[[i]] contains a list of parameters for method i. (Example: params[[i]]$id will give the identifier for the method used to label the output by the ClusterReport.)
`clusters`	A list of lists. clusters[[i]] contains a vector corresponding to the cluster assignments for the method specified in params[[i]].

Ted Laderas (laderast@ohsu.edu)

ClusterReport, clusterList, methodlistAuto, getClusterInfo, autoCompareClustering

  data(chocellcycle)
    ##run with input geometry of 5 and verbose output
    results <- CompareClustering(chocellcycle, verbose=TRUE, auto = TRUE, kinput=5)
    ##display results
    ClusterReport(results)