Options

different output from same oprator in Rapidminer, for forward and backward feature selection

KianKian Member Posts: 8 Contributor I
edited December 2018 in Help

Hi I want to doing feature selection with rapidminer with forward selection and backward elimination. there are three operator: 1:optimize selection(forward and backward), 2: optimize weight forward and backward and 3: forward selection and backward elimination operators. all there operator needs to have a inner operator like cross-validation operator toe evaluate the model. But in output of these three operator there are different selected feature and different accuracy. also when i do a cross validation separately with selected attributes by select attribute operator without using the forward and backward operator and with only use the their selected attributes, there are different results. i dont know whats wrong here! anyone can help? thanks

Answers

  • Options
    sgenzersgenzer Administrator, Moderator, Employee, RapidMiner Certified Analyst, Community Manager, Member, University Professor, PM Moderator Posts: 2,959 Community Manager

    hello @Kian welcome to the community! I'd recommend posting your XML process here (see https://youtu.be/KkgB5QXWXJ8 and "Read Before Posting" on right when you reply) and attach your dataset. This way we can replicate what you're doing and help you better.

    Scott

  • Options
    MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,508 RM Data Scientist

    Dear Kian,

     

    there is nothing wrong with it. Forward and Backward Selection have to deliver different results. 

     

    Best,

    Martin

    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
  • Options
    IngoRMIngoRM Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, Community Manager, RMResearcher, Member, University Professor Posts: 1,751 RM Founder

    If you want to learn more about why they typically deliver different results, I would recommend to check out this blog post here: https://community.rapidminer.com/t5/RapidMiner-Studio-Knowledge-Base/Multi-Objective-Feature-Selection-Part-1-The-Basics/ta-p/45775

     

    Cheers,

    Ingo

  • Options
    sgenzersgenzer Administrator, Moderator, Employee, RapidMiner Certified Analyst, Community Manager, Member, University Professor, PM Moderator Posts: 2,959 Community Manager

    nice link, @IngoRM :)

     

    Scott

     

     

Sign In or Register to comment.