Microsoft Certified Azure Fundamentals · Free Practice Question Medium
Question 56
You are working on an Azure Machine Learning designer project which involves creating an experiment that cleans and transforms data, trains a model, and then scores a model. You realize that your experiment's performance is sub-optimal. What could be potential ways to improve the performance of your experiment?
Select only one answer.
-
A
Use a larger dataset.
-
B
Increase the iteration count.
-
C
Decrease the number of features in the dataset.
-
D
Modify the experiment to use a different machine learning algorithm.
Reveal correct answer
Correct answer: D
A.
Simply using a larger dataset doesn't guarantee performance improvement, it may even degrade performance due to increased complexity.
B.
Increasing the iteration count could improve model accuracy, but it's not necessarily guaranteed to improve the experiment's performance.
C.
Decreasing the number of features might make the model simpler and faster, but it may also decrease the model's predictive power.
D.
Switching to a more suitable machine learning algorithm can potentially improve the performance of the experiment.
Discussion
Think the marked answer is wrong, or have a better explanation? Share it below — comments appear after review.
