RapidMiner 9.7 is Now Available

Lots of amazing new improvements including true version control! Learn more about what's new here.


[SOLVED] Has RM concept of metadata associated to ExampleSet, not attribute?

owenowen Member Posts: 22 Contributor II
edited November 2018 in Help
Hello all,

In RM is there concept of metadata associated to ExampleSet (not to attribute)?

For example, I have conducted an experiment on three subjects "Joe", "Jane", "Fred", and there are three datafiles that have the same format, but different content. I have a file reader that converts each file to an ExampleSet.

So after I load the files into RM, I have exampleSet1 from experiment1, exampleSet2 from experiment2, exampleSet3 from experiment3 and so on.
What RM abstraction helps me associate "Joe" to exampleSet1 , "Jane" to examplSet2 etc.?

I suppose I could add another attribute "Subject Name" to the exampleSet, but then I have a strange sparsity situation.
I might have 1e5 examples with numerical data and "Subject Name" not set.
And one example with no numerical data, and "Subject Name" set.
Is this the right way to go about it?

In the end result, I suppose, this metadata might be used for labelling charts and plots. E.g. blue curve is "Fred", red curve is "Joe" etc.

Thank you for your advice.



  • Marco_BoeckMarco_Boeck Team Lead Software Engineering Administrator, Moderator, Employee, Member, University Professor Posts: 1,938   RM Engineering

    have a look at the "Annotate" operator which can be used to, well, annotate things on your data. These annotations can be seen in the results perspective under the "Annoations" tab and they can be used with the other Annotations operators.

  • owenowen Member Posts: 22 Contributor II
    Thank you!
Sign In or Register to comment.