🦉 🎤   RapidMiner Wisdom 2020 - CALL FOR SPEAKERS   🦉 🎤

We are inviting all community members to submit proposals to speak at Wisdom 2020 in Boston.


Whether it's a cool RapidMiner trick or a use case implementation, we want to see what you have.
Form link is below and deadline for submissions is November 15. See you in Boston!

CLICK HERE TO GO TO ENTRY FORM

How to weight examples like SPSS?

tmyerstmyers Member Posts: 21 Contributor I
edited November 2018 in Help
Hi. I am new to RM and am trying to learn about its Weighting functions. I want to weight a dataset like in SPSS, where I select a single weight variable to weight examples (or cases in SPSS-speak). I would select the variable to use as the weight, and any subsequent aggregations I would perform would apply this weight variable to the examples included in the calculation.

I have been reading the definitions of the various Weighting variables and most seem to either derive weights from the dataset itself OR apply weights on an attribute-by-attribute basis......I'm just just looking to apply a single weight variable to everything.

Any advice would be greatly appreciated-

Thanks,

Tim
Tagged:

Answers

  • mschmitzmschmitz Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 2,156  RM Data Scientist
    Hi tim,

    you can set a attribute to the role weight using Set Role operator. This is then used in models, aggregations on an example-by example basis. Of course you can first generate the attribute with a Generate Attributes operator.

    ~Martin
    - Head of Data Science Services at RapidMiner -
    Dortmund, Germany
  • tmyerstmyers Member Posts: 21 Contributor I
    Thanks Martin. That seems to work for me so long as I remove any examples with a NULL value for the attribute set as the Weight role. I had a few stray examples that didn't have a value for my Weight attribute.

    Am I correct in assuming that whichever attribute I set as a Weight must have values for all examples?
Sign In or Register to comment.