Same-size k-Means

This component identifies equally-sized clusters of homogeneous items in a dataset. First, K-means algorithm is used to derive centers of the clusters. Then, every point is associated in turn to the closest cluster, in a way that forces each cluster to be of equal size. The size of each cluster will approximately be the number of original rows divided by the number of clusters.

The clustering algorithm uses the Euclidean distance on the selected attributes. The data is not normalized by the node (if required, you should consider to use the "Normalizer" as a preprocessing step).

This component is free to use and modify.
Author: Andrea De Mauro, aboutbigdata.net

Options

Columns selection (all numeric columns will be used for clustering)

Select columns to be kept (only and all numeric columns will be used for clustering).

Number of clusters

Number of clusters to be created.

Nodes

• Column Filter4 ×
• Java Edit Variable (simple)3 ×
• Math Formula2 ×
• RowID2 ×
• Sorter2 ×
• Chunk Loop Start1 ×
• Column Appender1 ×
• Column Filter Configuration1 ×
• Column Rename1 ×
• Component Input1 ×
• Component Output1 ×
• Constant Value Column1 ×
• Duplicate Row Filter1 ×
• Integer Configuration1 ×
• Loop End (Column Append)1 ×
• Numeric Distances1 ×
• Recursive Loop End1 ×
• Recursive Loop Start1 ×
• Row Splitter1 ×
• Similarity Search1 ×
• String Manipulation1 ×
• String To Number1 ×
• Table Row to Variable1 ×
• k-Means1 ×

Extensions

• FeatureKNIME Base nodes
• FeatureKNIME Distance Matrix
• FeatureKNIME Javasnippet
• FeatureKNIME Math Expression (JEP)
• FeatureKNIME Quick Forms

Links

• No links available