Overview of SuperNodes

One of the reasons that the IBM® SPSS® Modeler visual programming interface is so easy to learn is that each node has a clearly defined function. However, for complex processing, a long sequence of nodes may be necessary. Eventually, this may clutter the stream canvas and make it difficult to follow stream diagrams. There are two ways to avoid the clutter of a long and complex stream:

• You can split a processing sequence into several streams that feed one into the other. The first stream, for example, creates a data file that the second uses as input. The second creates a file that the third uses as input, and so on. You can manage these multiple streams by saving them in a project. A project provides organization for multiple streams and their output. However, a project file contains only a reference to the objects it contains, and you will still have multiple stream files to manage.
• You can create a SuperNode as a more streamlined alternative when working with complex stream processes.

SuperNodes group multiple nodes into a single node by encapsulating sections of a data stream. This provides numerous benefits to the data miner:

• Streams are neater and more manageable.
• Nodes can be combined into a business-specific SuperNode.
• SuperNodes can be exported to libraries for reuse in multiple data mining projects.