Auto Model one hot encoding

Chemical_engChemical_eng Member Posts: 5 Contributor I
Hello, I am using automodel in rapid miner studio. The problem I have is that I have a categorical variable, therefore one hot encoding is needed. However, the number of categories is more than 50 , what can I do so the algorithm takes it into consideration and does not ignore it 
Tagged:

Answers

  • mschmitzmschmitz Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,249 RM Data Scientist
    Hi,
    you can use the operator Replace Rare to replace Rare values with OTHER.
    BR,
    Martin
    - Head of Data Science Services at RapidMiner -
    Dortmund, Germany
  • mschmitzmschmitz Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,249 RM Data Scientist
    Oh, alternatively you can use Target Encoding of course.
    - Head of Data Science Services at RapidMiner -
    Dortmund, Germany
  • Chemical_engChemical_eng Member Posts: 5 Contributor I
    Hi , thanks, I was using auto model in studio , I guess for this I need to export it into a process, where in the process I would need to add this ? thanks 
  • mschmitzmschmitz Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,249 RM Data Scientist
    Hi,
    yes, or just do it up front with a small prep process. Please shoot me or Vlad an e-mail if you want to do this together with us.

    BR,
    Martin
    - Head of Data Science Services at RapidMiner -
    Dortmund, Germany
Sign In or Register to comment.