Google Professional Machine Learning Engineer Exam - Topic 11 Question 55 Discussion

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

Question #: 55
Topic #: 11

[All Professional Machine Learning Engineer Questions]

You work for a retailer that sells clothes to customers around the world. You have been tasked with ensuring that ML models are built in a secure manner. Specifically, you need to protect sensitive customer data that might be used in the models. You have identified four fields containing sensitive data that are being used by your data science team: AGE, IS_EXISTING_CUSTOMER, LATITUDE_LONGITUDE, and SHIRT_SIZE. What should you do with the data before it is made available to the data science team for training purposes?

ATokenize all of the fields using hashed dummy values to replace the real values.

BUse principal component analysis (PCA) to reduce the four sensitive fields to one PCA vector.

CCoarsen the data by putting AGE into quantiles and rounding LATITUDE_LONGTTUDE into single precision. The other two fields are already as coarse as possible.

DRemove all sensitive data fields, and ask the data science team to build their models using non-sensitive data.

Show Suggested Answer

Suggested Answer: A

by Casie at Nov 21, 2022, 02:49 AM

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

Mitsue

4 months ago

Definitely need to hash those values for security!

upvoted 0 times

...

Nilsa

4 months ago

Coarsening data could work, but is it enough?

upvoted 0 times

...

Lore

4 months ago

Wait, can PCA really protect sensitive info? Sounds risky.

upvoted 0 times

...

Chau

4 months ago

Disagree, removing all sensitive data seems too extreme!

upvoted 0 times

...

Shaniqua

4 months ago

I think tokenizing is the safest bet here.

upvoted 0 times

...

Iluminada

5 months ago

Removing all sensitive fields sounds extreme, but I guess it could be a safe option if we have enough non-sensitive data to work with.

upvoted 0 times

...

Adelle

5 months ago

Coarsening AGE and LATITUDE_LONGITUDE makes sense, but I'm not confident if it's enough to protect sensitive data.

upvoted 0 times

...

France

5 months ago

PCA seems like it could help, but I think it might lose important information about the original data.

upvoted 0 times

...

Fannie

5 months ago

I remember we discussed tokenization in class, but I'm not sure if hashing is the best approach for all these fields.

upvoted 0 times

...

Robt

5 months ago

Wait, I thought the expenditure approach also included things like transfer payments and money supply. This question is tripping me up a bit. I'll have to review my notes to make sure I understand this properly.

upvoted 0 times

...

Cherry

5 months ago

This question seems straightforward, I think I can handle it.

upvoted 0 times

...

Franklyn

5 months ago

Okay, let's think this through step-by-step. We need a virtual receptionist to connect callers, and a call queue to handle the help desk calls in FIFO order. I'm pretty confident those are the right resources to create.

upvoted 0 times

...