Exam AWS Certified Machine Learning - Specialty All Questions

View all questions & answers for the AWS Certified Machine Learning - Specialty exam

Exam AWS Certified Machine Learning - Specialty topic 1 question 277 discussion

Exam question from Amazon's AWS Certified Machine Learning - Specialty

Question #: 242
Topic #: 1

[All AWS Certified Machine Learning - Specialty Questions]

A company is planning a marketing campaign to promote a new product to existing customers. The company has data for past promotions that are similar. The company decides to try an experiment to send a more expensive marketing package to a smaller number of customers. The company wants to target the marketing campaign to customers who are most likely to buy the new product. The experiment requires that at least 90% of the customers who are likely to purchase the new product receive the marketing materials.

The company trains a model by using the linear learner algorithm in Amazon SageMaker. The model has a recall score of 80% and a precision of 75%.

How should the company retrain the model to meet these requirements?

A. Set the target_recall hyperparameter to 90%. Set the binary_classifier_model_selection_criteria hyperparameter to recall_at_target_precision.
B. Set the target_precision hyperparameter to 90%. Set the binary_classifier_model_selection_criteria hyperparameter to precision_at_target_recall.
C. Use 90% of the historical data for training. Set the number of epochs to 20.
D. Set the normalize_label hyperparameter to true. Set the number of classes to 2.

Show Suggested Answer

Suggested Answer: A 🗳️

by ADVIT at July 8, 2023, 6:42 a.m.

Disclaimers:

- ExamTopics website is not related to, affiliated with, endorsed or authorized by Amazon.
- Trademarks, certification & product names are used for reference only and belong to Amazon.

Comments

Submit Cancel

endeesa

7 months ago

Selected Answer: A

This is simple, assume we identify 100 customers who have spent atleast $1M dollars before. We want our marketing materil to reach atleast 90 of them - hence as many true positives as possible meaning FPs are also ok. This means we need to optimise for recall

upvoted 2 times

...

DimLam

8 months ago

Selected Answer: A

For me it is A. We need to send the promo to as many potential buyers as possible. so we need to reduce FN

upvoted 3 times

...

goku58

9 months, 4 weeks ago

Selected Answer: A

FPs should be fine as it is not mentioned in the question that it is expensive to send the expensive marketing packages to some extra customers. BUT FNs more than 10% are not fine as they want at least 90% of potential customers to get the material. Hence we tune hyperparameters to increase recall to target 90%. And then find best precision with this 90% recall. Which makes precision_at_target_recall an appropriate hyperparameter to set. But option A has recall_at_target_precision. Still leaning towards A as target recall needs to be 90%. As per docs, "If binary_classifier_model_selection_criteria is recall_at_target_precision, then precision is held at this value while recall is maximized."

upvoted 1 times

...

ashii007

10 months ago

we want to reduce FPs. Hence precision target should be 90%. Thus B.

upvoted 1 times

...

boledadian

10 months ago

Selected Answer: B

B We want to reduce the false positive(We do not want to give the expensive marketing package to the costomer who won't buy our product)

upvoted 3 times

DimLam

8 months ago

But I see that the goal is to send the promo to as much potential purchasers as possible (reduce FN)

upvoted 1 times

...

kaike_reis

10 months, 1 week ago

Selected Answer: A

(A) We want to reduce FN as much as possible (customers who will buy, but who were predicted by the model as “non-buyers”), so the correct alternative is Letter A.

upvoted 1 times

kaike_reis

10 months, 1 week ago

c - d are decoy.

upvoted 1 times

...

Mickey321

10 months, 4 weeks ago

Selected Answer: A

For classification problems, such as the one here, you can specify the binary_classifier_model_selection_criteria hyperparameter to control how the best model is selected from the validation set. You can choose from several criteria, such as accuracy, precision, recall, F1 measure, or cross-entropy loss1.

upvoted 2 times

...

ADVIT

11 months, 3 weeks ago

I think it's A - Recall.

upvoted 2 times

...