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

Combined labels

mansourmansour Member Posts: 26 Contributor II
Dear All
I would like to run attribute weighting models on a dataset with 5 features as labels. I want to combine the features (not running the model for each label every time, I mean I don't want to use the loop label operator). I want the see the weight for each feature based on 5 label combinations.
Any suggestion?
Many thanks.
Mansour

Answers

  • Options
    jmerglerjmergler Administrator, Moderator, Employee, RapidMiner Certified Analyst, Member, University Professor Posts: 41 Guru
    Hi Mansour
    Are you able to provide a little more detail or sample of the data? Are these numeric or categorical labels? 
    Jeff
  • Options
    mansourmansour Member Posts: 26 Contributor II
    Hi @jmergler
    Thanks for your reply. They are both numerical and categorical and in some instances a mixture of both.
    Best wishes.
    Mansour
  • Options
    jmerglerjmergler Administrator, Moderator, Employee, RapidMiner Certified Analyst, Member, University Professor Posts: 41 Guru
    Hi @mansour
    I think for the purposes of seeing the weights across all labels, one approach would be to bin the numeric labels as necessary, and simply concatenate your labels. You end up with a high number of unique classes, but it may serve your purposes, particularly if you have enough data to support it.  If you are able to provide any more information about the data, like a sample, and the size, and any process that you might have, then we might be able to get some more responses.
    Jeff
  • Options
    mansourmansour Member Posts: 26 Contributor II
    Hi @jmergler
    Many thanks. To make it simple, I converted all numeric label variables to Quartiled (so there are four classes in each variable, Q1 to Q4). Now how can I compute weights for each regular feature based on five categorical target features (not individually). 
    Best wishes.
    Mansour
Sign In or Register to comment.