transforming categorical values to dummy

uzjpkuzjpk Member Posts: 3 Contributor I
edited November 2018 in Help


 I am a student and for one of my projects in predicting house prices, I want to use regression method to predict the house prices. But as usual there are some categorical attributes in my data sets, which I would like to transform them to dummy variables (there are 3 different types of data type in my attributes, namely; integer, polynomial and binominal) before any further action. I was wondering whether you could help me in this regard and tell me how and with what operator can i to transform each of these attributes to dummy variables?



with best regards


  • Options
    Telcontar120Telcontar120 Moderator, RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 1,635 Unicorn

    You actually have several options here, depending on exactly what you want to do.

    "Nominal to Numerical" will allow you to create binary/dummy variables out of each nominal category present when you select coding type "dummy coding," although if you want to put them all into a classic linear regression you may prefer the "effect coding" which will omit one reference category that you select, so you don't end up with a set of perfectly collinear predictors.  

    You can also accomplish similar things with numerical variables by first turning them into categorical variables by binning them using one of the many binning operators, and then using the "Nominal to Numerical" operator after that.

    I hope this helps.




    Brian T.
    Lindon Ventures 
    Data Science Consulting from Certified RapidMiner Experts
  • Options
    uzjpkuzjpk Member Posts: 3 Contributor I

    thank you for your quick reply ,i'll try that. :)


  • Options
    tamararidwantamararidwan Member Posts: 7 Contributor I
    @uzjpk hi, i'm wondering if you can share your work with me? since i have the same projects and a student too, but i'm new in rapidminer. it would be great if you share the house price prediction. thx
Sign In or Register to comment.