Treemap node

This node/feature is available only if it is enabled in the PolyAnalyst Server license.

The Treemap node generates a treemap from the input dataset. A treemap is a visualization tool that displays dimensions of a dataset in a graphical way so that certain patterns in the hierarchical (aggregated) data become more apparent.

Introduction to treemaps

A treemap is a modern method of visualizing data that is particularly intended for hierarchical data. A collection of folders on your computer, with each folder containing subfolders, is an example of the hierarchical data. In this case, if you were examining a collection of files that are organized into folders, the collection of files, together with the containing folder information, would constitute a hierarchical dataset.

Treemaps excel at making patterns within the hierarchical data more apparent. The treemap is very similar to certain types of OLAP visualizations (see the separate section for details).

A treemap displays a collection of nodes, or groups of values, in the form of rectangles. The size and color of each rectangle corresponds to a particular dimension (column) of the input dataset. If there is a hierarchical relationship between two dimensions, dependent nodes are displayed within the nodes of the upper level.

Here is a trivial example. Suppose you have collected some information about the files on your computer. In the dataset, each file is represented by a separate record and file properties are represented by columns. There are also several key columns: the path to the file from the root folder, the size of the file, and the indication of the file type. You could input this trivial dataset into a Treemap node to generate a treemap that depicts the collection of files. You might find interesting patterns, e.g. you can easily single out folders that contain multiple large files.

The value of the treemap lies in the fact that this visualization tool is relatively adept at displaying a very large number of items on the screen in a compact space, and that certain patterns become readily apparent, particularly when there is a correlation to the hierarchical nature of the items present in the data.

You can be creative in deciding what constitutes a hierarchy. Nearly any set of categorical/nominal properties in the input dataset can be used for aggregation to form a hierarchy.

Flowchart connections

The Treemap node requires a single incoming connection with the node that generates a dataset, such as a data source, row, column or table operation node.

Like most visualization nodes, the Treemap node is a terminal node that generates a chart as a report. The node can be used as a parent to an Export to Image node (to save the chart to a PNG or PDF file) and Generic Dataset node (to present results in the tabular form).