SOMB {MLInterfaces} | R Documentation |
Self-organizing maps are covered in various machine learning
reviews. These can be viewed as clustering procedures
analogous to kmeans
but with a constraint
on the complexity of the subspace in which the cluster-defining
prototypes are identified.
The outputs of som
and SOM
are
not easily formulated as clustOutput-class
instances,
so classes that extend list
are used as output containers.
SOMB(exprObj, classifLab) somB(exprObj, classifLab)
exprObj |
An instance of the ExpressionSet class. |
classifLab |
A vector of class labels. |
You may pass additional parameters, see SOM
or som
for details.
An object of class SOMBoutput
or somBoutput
Jess Mar, VJ Carey <stvjc@channing.harvard.edu>
# access and trim an ExpressionSet library(golubEsets) data(Golub_Merge) smallG <- Golub_Merge[1:60,] # set a PRNG seed for reproducibilitiy set.seed(1234) # needed for nnet initialization # now run the classifiers #somB( smallG, "ALL.AML", 3) SOMB( smallG, "ALL.AML", 3)