Evaluate by cross-validation (leave-one-out) the effect induced by leaving out each performance on the result of a functional clustering.
ftest_plot_performances(fres, rtest,
main = "Title", opt.crit = "Jaccard", opt.perf = NULL)an object resulting from a functional clustering
obtained with the whole dataset using the function fclust.
a list of matrices,
each containing the results for a clustering index.
rtest is an object generated by the function ftest.
a string, that is used as the first, reference part of the title of each graph.
a list of strings,
indicating the clustering indices to plot.
The indices can be:
"Czekanowski_Dice", "Folkes_Mallows", "Jaccard", "Kulczynski",
"Precision", "Rand", "Recall", "Rogers_Tanimoto", "Russel_Rao",
"Sokal_Sneath1" or "Sokal_Sneath2".
For more informations, see the notice of R-package clusterCrit.
a list, that can include
a list, that can include
opt.comp = list("all.together", "performances.together",
"sorted.leg").
This option list manages the plot
of results obtained using the function ftest
with opt.var = "performances".
The item order in list is any.
"all.together", "performances.together"
plot (i) the general mean index;
(ii) the mean indices for each removed performance on a same plot,
when removing one after one each performance from the dataset.
This allows to evaluate the raw robustness of functional clustering
to perturbation of dataset,
and the weight of each performance on functional clustering.
"sorted.leg" plot
the names of performances decreasingly sorted
according to their weight on functional clustering.
"all" plot all possible graphs.
This option is equivalent to
opt.comp = list("all.together", "performances.together",
"sorted.leg").
Nothing. It is a procedure.
The trees obtained by leaving out each performance
are compared to the reference tree obtained with all performances
using different criteria :
"Czekanowski_Dice", "Folkes_Mallows", "Jaccard", "Kulczynski",
"Precision", "Rand", "Recall", "Rogers_Tanimoto", "Russel_Rao",
"Sokal_Sneath1" and "Sokal_Sneath2" index.
For more informations, see the notice of R-package clusterCrit.
Package "clusterCrit": Clustering Indices, by Bernard Desgraupes (University of Paris Ouest - Lab Modal'X)