kmeans: K-Means Clustering. 'kmeans' performs k-means clustering on...

In Fuzzy-Logix/AdapteR: Converts R syntax to SQL to provide fast analytics using in-database C++ library DB Lytix.

Description

The DB Lytix function called is FLKMeans.K-Means clusters the training data. The relationship of observations to clusters has hard edges. In general, k-means has two steps: assigning data points to the nearest cluster and then moving each cluster’s centroid to the center of the members of the cluster.

Usage

kmeans(x, ...)

Arguments

x	an object of class FLTable, wide or deep
centers	the number of clusters
iter.max	the maximum number of iterations allowed
nstart	the initial number of random sets
excludeCols	the comma separated character string of columns to be excluded
classSpec	list describing the categorical dummy variables
whereconditions	takes the where_clause as a string

Value

kmeans returns a list which replicates equivalent R output from kmeans in stats package. The mapping table can be viewed using object$mapping if input is wide table.

Constraints

If classSpec is not specified, the categorical variables are excluded from analysis by default.

See Also

kmeans for R reference function implementation.

Examples

widetable  <- FLTable(getTestTableName("tblAbaloneWide"), "ObsID", whereconditions = "ObsID < 500")
kmeansobject <- kmeans(widetable,3,20,1,"Rings,SEX")
print(kmeansobject)
plot(kmeansobject)
One can specify ClassSpec and transform categorical variables 
before clustering. This increases the number of variables in the plot
because categorical variable is split into binary numerical variables.
The clusters may not be well-defined as is observed in the case below:-
kmeansobjectnew <- kmeans(widetable,3,20,1,"Rings,SEX",list("DummyCat(D)","SEX(M)"))
plot(kmeansobjectnew)

Fuzzy-Logix/AdapteR documentation built on May 6, 2019, 5:07 p.m.