Amazon MLS-C01 Exam - Topic 5 Question 74 Discussion

Actual exam question for Amazon's MLS-C01 exam

Question #: 74
Topic #: 5

A finance company has collected stock return data for 5.000 publicly traded companies. A financial analyst has a dataset that contains 2.000 attributes for each company. The financial analyst wants to use Amazon SageMaker to identify the top 15 attributes that are most valuable to predict future stock returns.

Which solution will meet these requirements with the LEAST operational overhead?

AUse the linear learner algorithm in SageMaker to train a linear regression model to predict the stock returns. Identify the most predictive features by ranking absolute coefficient values.

BUse random forest regression in SageMaker to train a model to predict the stock returns. Identify the most predictive features based on Gini importance scores.

CUse an Amazon SageMaker Data Wrangler quick model visualization to predict the stock returns. Identify the most predictive features based on the quick model's feature importance scores.

DUse Amazon SageMaker Autopilot to build a regression model to predict the stock returns. Identify the most predictive features based on an Amazon SageMaker Clarify report.

Show Suggested Answer

Suggested Answer: B, D

https://www.redhat.com/en/topics/devops/what-is-blue-green-deployment

by Georgene at Apr 25, 2023, 10:43 PM

Limited Time Offer

25%

Off

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

Contribute your Thoughts:

Submit Cancel

Cecily

4 months ago

I disagree, C could be useful for quick insights without much hassle.

upvoted 0 times

...

Launa

4 months ago

Surprised that no one mentioned feature selection techniques!

upvoted 0 times

...

Shawna

4 months ago

A linear model? Really? Seems too simplistic for this data.

upvoted 0 times

...

Samira

4 months ago

I think B is solid too, random forests are powerful!

upvoted 0 times

...

Cammy

4 months ago

Option D seems like the best choice for low overhead.

upvoted 0 times

...

Hubert

5 months ago

I recall that using Data Wrangler can simplify the workflow, but I'm uncertain if it provides the best feature importance compared to the other options.

upvoted 0 times

...

Marla

5 months ago

I practiced a similar question about feature selection, and I feel like the random forest method might be more robust, but it could be more complex to set up.

upvoted 0 times

...

Merrilee

5 months ago

I think using Amazon SageMaker Autopilot could be a good choice since it automates a lot of the process, but I need to double-check how it handles feature importance.

upvoted 0 times

...

Lou

5 months ago

I remember studying about feature importance in regression models, but I'm not sure which method has the least overhead.

upvoted 0 times

...

Merlyn

5 months ago

Hmm, I'm a bit unsure about this. I know the Orchestrator is some kind of management component, but I'm not sure if it's specifically a load balancer or network switch. I'll have to think this through a bit more.

upvoted 0 times

...

Benton

5 months ago

Okay, I've got this. Collecting patient satisfaction data is a common way to evaluate medical students' performance and bedside manner. That's got to be the right answer here.

upvoted 0 times

...

Loren

9 months ago

Ah, the age-old question: linear or random forest? It's like choosing between a cup of coffee or a Red Bull - both will get the job done, just depends how much of a caffeine kick you want.

upvoted 0 times

Tiffiny

8 months ago

Sounds like a plan, it's always good to try different models to see which one performs better.

upvoted 0 times

...

Yoko

8 months ago

That's a good point, overfitting can be a problem with random forest. Maybe start with linear regression and then try random forest to compare results.

upvoted 0 times

...

Lou

9 months ago

True, random forest can handle non-linear relationships better, but it may overfit the data.

upvoted 0 times

...

Fallon

9 months ago

Linear regression is simpler and easier to interpret, but random forest may capture more complex relationships in the data.

upvoted 0 times

...

Hector

10 months ago

Options A and B are like the finance equivalent of rock-paper-scissors. I'd go with whichever one makes the most intuitive sense to me.

upvoted 0 times

...

Nieves

10 months ago

Option D with Amazon SageMaker Autopilot sounds like the easiest solution, but I'm curious to see how the Clarify report compares to the other methods. Might be worth exploring if I'm feeling lazy.

upvoted 0 times

Leonora

9 months ago

Let's give Autopilot a try and see how the Clarify report stacks up against the other options. It could save us a lot of time and effort.

upvoted 0 times

...

Pok

9 months ago

I agree, the Clarify report from Autopilot will give us valuable insights into the most predictive features for stock returns.

upvoted 0 times

...

Corinne

9 months ago

Option D with Amazon SageMaker Autopilot is definitely the way to go. It's the easiest and most efficient method.

upvoted 0 times

...

Erin

10 months ago

Option C looks interesting, but I'm not familiar with the quick model visualization in SageMaker. I'd need to do some research to see how reliable the feature importance scores are for this use case.

upvoted 0 times

Detra

8 months ago

I agree, Option D with Amazon SageMaker Autopilot could be a good option. The Clarify report could help identify the most predictive features.

upvoted 0 times

...

Gerri

8 months ago

I think Option B could also be effective. Gini importance scores from random forest regression might provide valuable insights.

upvoted 0 times

...

Margart

9 months ago

Option A seems like a solid choice. Using linear regression to rank coefficient values could help identify the top attributes.

upvoted 0 times

...

Avery

10 months ago

I'm leaning towards Option B. Random forests can handle nonlinear relationships and the Gini importance scores should give me a good sense of the top predictive features.

upvoted 0 times

Kirk

9 months ago

I agree, Option B seems like the best option for identifying the most valuable attributes for predicting stock returns.

upvoted 0 times

...

Stevie

9 months ago

Gini importance scores can definitely help identify the top predictive features.

upvoted 0 times

...

Lucille

9 months ago

Option B sounds like a good choice. Random forests are great for handling nonlinear relationships.

upvoted 0 times

...

Holley

10 months ago

Option A seems like a good starting point. Linear regression is a classic approach for this type of problem, and ranking the coefficient values is a straightforward way to identify the most important features.

upvoted 0 times

...