RapidMiner 9.7 is Now Available

Lots of amazing new improvements including true version control! Learn more about what's new here.

CLICK HERE TO DOWNLOAD

Storing and Using Confidence values

fstarsinicfstarsinic Member Posts: 20 Contributor II
edited May 8 in Help
I've created several models that work fairly well.
They return confidence values, which are useful (because I can sort a class by confidence).
When I combine the models and use a voting classifier, the resultant confidence values are just the percentage of total votes and the granular values are gone.Β  I'm left with a far less granular set of confidence values, e.g., (0%,25%,50%,75%100%).

Is there a way to store the more granular values, combine them and get results that report the overall more granular confidence rather than just the vote percentage?

Tagged:

Answers

  • mschmitzmschmitz Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 2,462  RM Data Scientist
    Hi @fstarsinic,
    i think what you want to do is to use Stacking.

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