Is optimization possible in a multi-label problem?
For example, there is data with 7 variables, 2 of which are designated as prediction targets (labels), and for convenience, we will call them A and B (real type labels).
A and B predictions could be performed without problems using multi-label modeling operators.
In addition to this, what I want to do is a multi-label version of ‘Optimize’.
‘Optimize’ in Model Simulator is my favorite feature
When there was only one label, I could easily check the value of each independent variable that maximized or minimized the label value using the optimize function.
Is this feature also available in multi-label models?
First, I confirmed that the multi-label model is not connected to the model simulator.
Even if I don't use a model simulator,
When there are two or more labels, I want to derive the values of each independent variable that maximize or minimize the values while satisfying the conditions of the given labels.
1. The conditions of the two labels satisfy A>10 and B<30.
2. At the same time, A is maximized and B is minimized.
I want to know what the values of each of the five variables are.
Is there an operator that performs the above functions?
I tried referring to the 'Prescriptive Optimization' and 'Prescriptive Analytics' operators, but I couldn't figure out if it was not the right operator or if I didn't know how to do it.
Any simple process or small tips would be greatly appreciated.