I've a dataset of multiple categorical columns along with a target column that is continuous. Assume combination of categorical columns has a different range of values of target. Ex
Col1 - col2 - col3 - target(in ranges) A - B - C - ( 100 -1000) A - C - B - ( 1 - 100) B - B - C - ( 5 - 50)
Now, let's assume A -C - B has a value 200 in Test. This should be caught as Novelty by trained model. Question: Do we have to feed Novelty detection model (like LOF) for each category combination separately? Meaning, in above example,Do I have train the model thrice for each combination?