Google Professional Machine Learning Engineer Exam - Topic 7 Question 54 Discussion

Actual exam question for Google's Professional Machine Learning Engineer exam

Question #: 54
Topic #: 7

[All Professional Machine Learning Engineer Questions]

You work for a magazine publisher and have been tasked with predicting whether customers will cancel their annual subscription. In your exploratory data analysis, you find that 90% of individuals renew their subscription every year, and only 10% of individuals cancel their subscription. After training a NN Classifier, your model predicts those who cancel their subscription with 99% accuracy and predicts those who renew their subscription with 82% accuracy. How should you interpret these results?

AThis is not a good result because the model should have a higher accuracy for those who renew their subscription than for those who cancel their subscription.

BThis is not a good result because the model is performing worse than predicting that people will always renew their subscription.

CThis is a good result because predicting those who cancel their subscription is more difficult, since there is less data for this group.

DThis is a good result because the accuracy across both groups is greater than 80%.

Show Suggested Answer

Suggested Answer: C

by Vilma at Oct 22, 2022, 01:39 PM

Limited Time Offer

25%

Off

Get Premium Professional Machine Learning Engineer Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

Submit Cancel

Terry

5 months ago

82% for renewals is kinda low, right?

upvoted 0 times

...

Lacey

5 months ago

But shouldn't we expect higher accuracy for renewals?

upvoted 0 times

...

Sommer

6 months ago

I think it's impressive that they nailed the cancellations!

upvoted 0 times

...

Viva

6 months ago

99% accuracy on cancellations? Sounds too good to be true.

upvoted 0 times

...

Huey

6 months ago

90% renew their subscription, only 10% cancel.

upvoted 0 times

...

Queenie

6 months ago

I think option D sounds reasonable since both accuracies are above 80%, but I'm not confident if that alone makes it a good result.

upvoted 0 times

...

Stanton

6 months ago

I feel like option A could be misleading. Just because the renewal accuracy is lower doesn't mean it's a bad model overall.

upvoted 0 times

...

Krystal

6 months ago

I remember a practice question where we discussed the importance of accuracy in imbalanced datasets. Maybe option C is valid because predicting cancellations is harder?

upvoted 0 times

...

Cassandra

6 months ago

I'm not entirely sure, but I think option B might be right since the model should ideally do better than just guessing everyone renews.

upvoted 0 times

...

Donte

6 months ago

Disabling query folding on the long-running queries could be worth a shot. I'll make sure to test that out.

upvoted 0 times

...

Lera

6 months ago

I'm pretty confident I know the answer to this one. I think it's option B - Yes, creating a Read Replica of another Read Replica is supported.

upvoted 0 times

...

Tawna

6 months ago

Hmm, I'm a bit unsure about this one. The options seem similar, and I want to make sure I understand the differences between them before selecting an answer.

upvoted 0 times

...