How to generate synthetic data with RM?

sherlocksherlock Member Posts: 24 Contributor II

to avoid overfitting I would like to increase the dataset based on real data, maybe using GAN? Are there example processes available?
Or is Python required for that? Are there ways using Turbo Prep?


Best Answer

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,485 RM Data Scientist
    Solution Accepted
    GAN is not supported yet. You may use SMOTE or Build Simulation in operator toolbox.
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany


  • sherlocksherlock Member Posts: 24 Contributor II
    edited June 2022
    Thanks, for your quick answer, very helpful! I just realized that the "Smote Upsampling"-Operator doesn't offer all parameters described in the help (advanced parameters enabled). E.g. minoriy classe, auto detect minority class, upsampling size etc.
    Do I have an old version (just downloaded from marketplace) or anything to activate?

Sign In or Register to comment.