Selecting Tokenized Attributes

gn582
gn582 New Altair Community Member
edited November 2024 in Community Q&A
Hey guys,

Using Rapidminer for the first time for one of my university subjects. I'm trying to make a decision tree that uses the least_square criterion as it has a numerical attribute as the label.

The data deals with Youtube attributes and I am trying to tokenize the tag column (which I can successfully do) and then select each tag individually using the "Select Attribute" operator so that I can input the data into the decision tree and have the number of likes column be the label for the decision tree. 

The problem is when I process the tag data and tokenize it, I can only select the attribute "text" rather than each token individually. Is there a way around this or a way I can select each token to input into the data set?

Sorry if I worded it confusingly,

Thanks for the help!
Tagged:

Best Answer

  • lionelderkrikor
    lionelderkrikor New Altair Community Member
    Answer ✓
    Hi @gn582,

    If I good understand, you want select only the token attributes and exclude the "text" attribute ?

    In this case : 
     - select attribute filter type = single
     - select your "text" attribute
    AND
     - check invert selection



    Regards,

    Lionel

Answers

  • lionelderkrikor
    lionelderkrikor New Altair Community Member
    Answer ✓
    Hi @gn582,

    If I good understand, you want select only the token attributes and exclude the "text" attribute ?

    In this case : 
     - select attribute filter type = single
     - select your "text" attribute
    AND
     - check invert selection



    Regards,

    Lionel
  • gn582
    gn582 New Altair Community Member
    Exactly what I wanted! Thanks heaps Lionel, was stuck on that problem for a while and couldn't figure out why it wouldn't work. Saved me so much time!

    Thanks again 
  • lionelderkrikor
    lionelderkrikor New Altair Community Member
    You're welcome @gn582,

    Good luck for your study ! 

    Regards,

    Lionel