Amazon MLS-C01 Exam - Topic 9 Question 67 Discussion

Actual exam question for Amazon's MLS-C01 exam

Question #: 67
Topic #: 9

A company that manufactures mobile devices wants to determine and calibrate the appropriate sales price for its devices. The company is collecting the relevant data and is determining data features that it can use to train machine learning (ML) models. There are more than 1,000 features, and the company wants to determine the primary features that contribute to the sales price.

Which techniques should the company use for feature selection? (Choose three.)

AData scaling with standardization and normalization

BCorrelation plot with heat maps

CData binning

DUnivariate selection

EFeature importance with a tree-based classifier

FData augmentation

Show Suggested Answer

Suggested Answer: C, D, F

https://towardsdatascience.com/feature-selection-using-python-for-classification-problem-b5f00a1c7028#:~:text=Univariate%20feature%20selection%20works%20by,analysis%20of%20variance%20(ANOVA).&text=That%20is%20why%20it%20is%20called%20'univariate'

https://arxiv.org/abs/2101.04530

by Mireya at Dec 07, 2022, 09:49 PM

Limited Time Offer

25%

Off

Get Premium MLS-C01 Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

Submit Cancel

Lonna

4 months ago

1,000 features? That's a lot! How do they even start narrowing it down?

upvoted 0 times

...

Shonda

4 months ago

Univariate selection is a classic technique, can't go wrong there!

upvoted 0 times

...

Janna

4 months ago

Wait, why would you use data binning for feature selection? Seems off.

upvoted 0 times

...

Lucy

4 months ago

Definitely agree, feature importance with tree-based classifiers is a must!

upvoted 0 times

...

Irma

4 months ago

Correlation plots are super useful for this!

upvoted 0 times

...

Ma

5 months ago

Data scaling seems important, but I don't think it's directly related to feature selection. I might skip option A, but I'm leaning towards B, D, and E.

upvoted 0 times

...

Muriel

5 months ago

Feature importance with tree-based classifiers sounds familiar. I think it helps identify which features are most impactful for predictions. I would go with option E for sure.

upvoted 0 times

...

Tamar

5 months ago

I'm not entirely sure about univariate selection. I think it involves looking at each feature individually, but I can't recall the specifics. Maybe option D?

upvoted 0 times

...

Precious

5 months ago

I remember we discussed correlation plots in class, and they can really help visualize relationships between features. I think option B is a good choice.

upvoted 0 times

...

Cortney

5 months ago

Okay, let me see. The question mentions the subscriber key is in the _subscribers data view, so it's likely not a global unsubscribe list issue. I'm leaning towards the Auto Suppression List as the source.

upvoted 0 times

...

Pura

5 months ago

This seems pretty straightforward. I think the likely outcome of creating this type of chart is a quantitative analysis that will help management better understand the financial risks and potential impacts. I'll focus on clearly presenting the probability, impact, and financial details for each risk event.

upvoted 0 times

...

Pearlene

5 months ago

This seems straightforward enough. I'll carefully read through the options and select the two that I believe are true based on my knowledge of the address book audit log feature.

upvoted 0 times

...

Magnolia

5 months ago

Hmm, I'm a bit confused by this one. I'll need to think it through carefully to make sure I understand the implications of the question.

upvoted 0 times

...