The Altair Community is migrating to a new platform to provide a better experience for you. The RapidMiner Community will merge with the Altair Community at the same time. In preparation for the migration, both communities are on read-only mode from July 15th - July 24th, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here.
Options

"Regress SVM with numeric and nominal?"

noah977noah977 Member Posts: 32 Maven
edited May 2019 in Help
Hello,

I want to build an Regression SVM that will train on a "score".  The problem is that the data has both numeric and nominal data. 

For example:

Age: numeric
Favorite Color: (red,green,blue)  nominal
Favorite Food: (meat, chicken,fish) nominal
Weight: numeric
Calories per day: numeric
Postal Code: (90026, 90028, etc.)  Looks numeric but really is nominal

The actual data has about 30 features of which about 15 are nominal and 15 are numeric.

Any ideas on how to build the proper data set and model for a regression SVM?

Thanks!
Tagged:

Answers

  • Options
    landland RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 2,531 Unicorn
    Hi,
    your problem is to transfer nominal values into numerical ones. You could use nominal2numeric, but I think it would be better to binaryze it first. This means, every nominal value of a nominal attribute becomes a column: favourite color = red and favourite color = green ... and so on. The cell will contain a true if the nominal value was the associated value and false otherwise.
    You then could translate this by nominal2numeric into numerical values processing with the svm
    This method prevents you to put in some ordinal information, by associating colors with numbers (green = 0, red = 1, blue = 2) which aren't simply in the data.

    Greetings,
      Sebastian
Sign In or Register to comment.