agnes: Agglomerative Nesting

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

Description

agnes computes agglomeraive hierarchial clustering on FLTable objects.

Usage

agnes(x, ...)

Arguments

x	an object of class FLTable, can be wide or deep table
diss	logical if x is dissimilarity matrix. currently not used
metric	only "euclidean" distance supported currently
Stand	logical indicating if standardization should be done before calculating diss matrix
method	character. Allowed methods are "average", "single", "complete", "centroid"
par.method	currently not used and always 0
keep.diss	logicals indicating if the dissimilarities and/or input data x should be kept in the result
keep.data	logicals indicating if the dissimilarities and/or input data x should be kept in the result
trace.lev	integer specifying a trace level for printing diagnostics during the build and swap phase of the algorithm. currently always 0
maxit	maximum number of iterations
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
distTable	name of the in-database table having dissimilarity matrix or distance table

Details

The DB Lytix function called is FLAggClustering. In the initialization, each observation in the dataset belongs to its own cluster. In each iteration, agglomerative clustering would aggregate the two clusters that are nearest to each other, for which the distance is measured by the linkage method. This would continue until either the entire dataset belongs to one cluster or until the maximum number of iterations has been reached

Value

agnes returns a list and replicates equivalent R output from agnes in cluster package. The mapping table can be viewed using mapping component, if input is wide table.

Constraints

Plotting for large datasets takes longer time to fetch data. Error is thrown if results cannot be fetched. maxit should be more than no.of. observations for algorithm to reach completion. Error is thrown if algorithm does not reach completion or more than one cluster is formed at any step. If classSpec is not specified, the categorical variables are excluded from analysis by default.

See Also

agnes for R reference implementation.

Examples

deeptable  <- FLTable("tblUSArrests", "ObsID","VarID","Num_Val")
agnesobject <- agnes(deeptable,maxit=50)
print(agnesobject)
plot(agnesobject)

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

widetable  <- FLTable( "iris", "obsid")
agnesobjectnew <- agnes(widetable,maxit=500,classSpec=list("Species(setosa)"))
The below plot throws warnings!
plot(agnesobjectnew)

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