If the performance of a classification model on the test set (out-of-sample) err
ID: 3890583 • Letter: I
Question
If the performance of a classification model on the test set (out-of-sample) error is poor, you can just re-calibrate your model parameters to achieve a better model.
Yes
No
Suppose you derived a classification model. The error you obtained on the training set is low and the error on the test set is large. The model suffers from...
under-fitting the data
over-fitting the data
If your model under-fit the data (recall the general statement describing trees: Top is green AND Bottom is brown), introducing more features to make the model more complex will help.
True
False
Explanation / Answer
Question a
Yes,If the performance of a classification model on the test set (out-of-sample) error is poor, you can just re-calibrate your model parameters to achieve a better model.
Question b
Suppose you derived a classification model. The error you obtained on the training set is low and the error on the test set is large. The model suffers from over-fitting the data
Question c
If your model under-fit the data (recall the general statement describing trees: Top is green AND Bottom is brown), introducing more features to make the model more complex will help.
True