The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here

How can I change type of weight in feature selection algorithms?

maryam_raz21maryam_raz21 Member Posts: 9 Contributor I
edited November 2018 in Help

Hi. I'm new in rapidminer and I try feature selection algorithm such as backward eliminatiom and svm-RFE. The problem is that in the result the weight of feature is only 0,1 and it's not useful for me. How can I chang it to real?

Tagged:

Answers

  • kypexinkypexin RapidMiner Certified Analyst, Member Posts: 291 Unicorn

    Hi @maryam_raz21

     

    To get real weights, you should use one of the 'Weight by ...' operators; the feature selection operators you mentioned are acutually only selecting relevant features (hence assigning them weight = 1) and drop irrelevant (hence assigning them weight = 0).

    -- 
    Vladimir
    http://whatthefraud.wtf
  • maryam_raz21maryam_raz21 Member Posts: 9 Contributor I

    Thanks for reply:smileyhappy: . I have  another question . This feature selection such as SVM-RFE always selecting relevant or just in rapidminer They behave like this؟ in the papers that I studeid, they used svm-RFE for bioinformatics widely!

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee-RapidMiner, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,533 RM Data Scientist

    Hi,

     

    have you checked the feature selection extension for this?

     

    Cheers,

    Martin

    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
  • maryam_raz21maryam_raz21 Member Posts: 9 Contributor I

    yes. I installed feature selection extension 1.1.4 that included svm-rfe, ensemble feature selection and etc. in result all of feature's weight that were selected are 1.:smileyindifferent:

Sign In or Register to comment.