• Load data from RapidMiner datasources
• Load preexisting processes
• Load data (Int, Double or String values; from up to 4 inports) from KNIME (but inside RapidMiner node you can use other data formats too)
• Check the correctness of the process (typing errors, ...)
• Populate possible options where possible inside the configurator (like column names, ...)
• Execute the RapidMiner process, this can include IO (files, or network communication)
• Give the resulting ExampleSets (the purple circles) as a result to KNIME (up to 4 outports)
• Include the KNIME row ids as a column on demand (so Highliting is possible around RapidMiner)

RapidMiner viewer node

• Load data (Int, Double or String values) from KNIME
• Visualize using the various plotters available in RapidMiner
• You can use the various option available on the RapidMiner result's view interface
• You can save the figures