Hierarchical Cluster View

This view takes a hierarchical cluster tree and the input table that was used for the creating the clustering. It visualizes the cluster dendrogram and provides a distance plot across all levels of the hierarchy.

Input Ports

Icon
The hierarchical cluster tree that has been created by the Hierarchical Clustering (DistMatrix) node.
Icon
The data table that was output by the Distance Matrix Calculate node.

Output Ports

This node has no output ports

Views

Dendrogram/Distance View
  • Dendrogram: This view presents a dendrogram that illustrates the cluster hierarchy. The x-axis is nominal, showing each data point with its row ID. The y-axis represents the fusion distance. At the bottom of the chart, individual data points are displayed which can be considered clusters of size one. The lines indicate which clusters have been fused by the clustering algorithm. The height of each connection represents the distance between clusters, as defined by the linkage type. Clusters can be selected and hilited, with all contained subclusters hilited as well.
  • Distance: The distance plot shows the overall fit of the data to a given number of clusters. This view helps to identify an optimal number of clusters, as the curve will typically have a knee at the point where additional fusions will merge clusters that are well separated, thus having high fusion distance. The x-axis represents the number of fusions. Note that the number of clusters corresponds to the number of data points minus the number of fusions. The y-axis represents the fusion distance. Points cannot be hilited since distances relate to dendrogram height, not individual data points. The appearance tab allows you to adjust the view by hiding or displaying dots, changing line thickness, and adjusting dot size.

Workflows

Links

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.