ANNOUNCEMENT: RAPIDMINER 9.1 HAS JUST BEEN RELEASED!   PLEASE DOWNLOAD AND GIVE FEEDBACK. ENJOY AND HAPPY RAPIDMINING!   -- @sgenzer – Community Manager

Related Values

ad058ad058 Member Posts: 2 Contributor I
edited November 10 in Help

Hello,

 

i am working on recommender and i am at the first step and doing Data preperation. This is an example how my data looks like: Bildschirmfoto 2017-12-05 um 20.43.43.png

I wannted to make my Algorithm understand that the names with the same 'matchid' are related together. 

It's important because I want to know which Champions(Name) are related togehter and later, if I've just got one name, I want to predict the other one. 

 

Hope someone can help me. Thnaks! 

Tagged:

Best Answer

  • mschmitzmschmitz Posts: 1,833  RM Data Scientist
    Solution Accepted

    Hi,

     

    I think Pivot is your friend :)

     

    Best,

    Martin

    - Head of Data Science Services at RapidMiner -
    Dortmund, Germany

Answers

  • Telcontar120Telcontar120 Moderator, RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 883   Unicorn

    You can use the Aggregate operator to group by mattch id and aggregate counts of purchases, etc.   Then run your analysis on the dataset prepared in that way.

     

     

    Brian T.
    Lindon Ventures 
    Data Science Consulting from Certified RapidMiner Experts
    sgenzer
  • ad058ad058 Member Posts: 2 Contributor I

    @Telcontar120 Thanks for your help. It's not really clear for me how thise gonna be work, but i'm fine with pivot.

     

    @mschmitz Thanks a lot ;) 

    sgenzer
Sign In or Register to comment.