Nominal to Numeric

emceemouliemceemouli Member Posts: 2 Contributor I
edited November 2018 in Help
Hi,

I am doing a basic Linear Regression and using Nominal to Numeric operator to convert my String data to Numeric to be used in the Linear Regression operator. I used the dummy coding configuration in the Nominal to Numeric operator which creates additional attributes with values 0 or 1 and my prediction looks good with this. But towards the end, I want to convert this back to the original format of data where I have 1 column with the different values. Can you please let me know how I could do that....

I tried the unique integers configuration in the nominal to numeric function which does not add any additional attributes or columns but the prediction is not right...Any clarity or help in this regard would be greatly helpful.

Thanks,
Chandra

Answers

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,503 RM Data Scientist
    Hi,

    i think the most convenient solution is to join the old data set back on the predicted data set.

    ~Martin
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
  • emceemouliemceemouli Member Posts: 2 Contributor I
    Thanks
Sign In or Register to comment.