It would be good to add a table describing the actions of the preprocessing model generated by the discretization operators.
When a discretization model is created and then stored in the repository it is not obvious what transformation the model will perform on data. Only the attributes that are affected.
May I suggest something like this?
It would really help for making the preprocessing model more portable and easy to extract into reports explainable to the wider business.
(Current workaround is aggregate results of the discretization column, but it's clunky).
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.