I have a bunch of attributes that get tossed into a model. Attribute P comes from a human selecting something from a drop-down list, and is thought to be wrong 30% of the time. When the user chooses the wrong P then all the aggregations
that happen downstream are wrong, and this in turn affects the model I'm trying to build. I have several thousand distinct values of P.
Is there a way that I can build a model & query that will somehow flag the values for P that seem to be wrong and suggest the correct value for P ?
View Complete Post