Meaning of 'others' in variable importance inside Machine learning modelling in DSS

kautuk
kautuk Partner, Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 5 Partner

Hi,

I have built a binary classification model using RandomForest. After the model is built, I am getting few variable value as 'others' inside variable importance. I do not have any value as 'others' in these variables and hence, I wanted to know what value DSS categorizes as 'others' on running the model.

Thanks,

Kautuk

Answers

  • Muennighoff
    Muennighoff Dataiker, Registered Posts: 3 Dataiker

    Hey! Variable importance shows importances for individual features after preprocessing is applied. The preprocessing may add new columns.

    In your case, you are likely preprocessing features like SN_PN with Dummy encoding, which creates an Other category for column values that do not fall within the specified Max nb. of categories. Increasing Max nb. of categories in the preprocessing window prior to training should remove the Other.

    Hope this helps

Setup Info
    Tags
      Help me…