optimal number of groups

Calculate the optimal number of groups for a future Kmeans with classical intergoup / intragroup variance simulation. If the maximum number of groups declared <= 2, first calculate the maximum reasonable number of groups (can be very long) using a confirmatory logistic regression or dicriminant analysis for each kmeans simulation and retaining the number of groups generating the maximum leap in accuracy. In that last case the number of groups to span is firstly delimited from 2 to sqrt(nbrows)*3.

Options

Select variables: Select variables
Maximum number of groups: Maximum number of groups (effective only for varinter/vartot method)
check option: used if ngroup <= 2%%00010method to determine the maximum number of cluster (maximum reasonable fragmentation).%%00010

Input Ports

: table with numeric variables

Output Ports

: optimal number of groups to inject in a Kmeans algorithm

Nodes

Column Filter Configuration1 ×
Column Rename1 ×
Component Input1 ×
Component Output1 ×
Integer Configuration1 ×
Merge Variables1 ×
Python Script (1⇒1)1 ×
Single Selection Configuration1 ×

Extensions

FeatureKNIME Base nodes
FeatureKNIME Python 2 Integration (legacy)
FeatureKNIME Quick Forms