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 282 discussion

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

Question #: 282
Topic #: 1

[All AWS Certified Machine Learning - Specialty Questions]

A machine learning (ML) specialist at a retail company must build a system to forecast the daily sales for one of the company's stores. The company provided the ML specialist with sales data for this store from the past 10 years. The historical dataset includes the total amount of sales on each day for the store. Approximately 10% of the days in the historical dataset are missing sales data.

The ML specialist builds a forecasting model based on the historical dataset. The specialist discovers that the model does not meet the performance standards that the company requires.

Which action will MOST likely improve the performance for the forecasting model?

A. Aggregate sales from stores in the same geographic area.
B. Apply smoothing to correct for seasonal variation.
C. Change the forecast frequency from daily to weekly.
D. Replace missing values in the dataset by using linear interpolation.

Show Suggested Answer

Suggested Answer: D 🗳️

by aquanaveen at Dec. 17, 2023, 12:49 p.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

taustin2

Highly Voted 1 year ago

Selected Answer: D

Could be B or D. The question calls out that 10% of the data is missing, which a lot. Smoothing would help as well. I'll go with D.

upvoted 9 times

...

ef12052

Most Recent 2 months, 3 weeks ago

Selected Answer: D

Linear interpolation is a widely used imputation method for time series data, where missing values are replaced by values calculated based on adjacent data points.

upvoted 1 times

...

AIWave

9 months, 3 weeks ago

Selected Answer: B

While both B & D will have effect on performace but MOST effect will be from B - smoothening of seasonal variations for forecasting. Linera interpolation may even have adverse effect on performance if relationship between variables in not linear.

upvoted 4 times

...

vkbajoria

9 months, 4 weeks ago

Selected Answer: B

smoothing is more important than missing data in this scenario

upvoted 2 times

vkbajoria

9 months ago

After must consideration, I will change my answer to "D" First and foremost, solving missing data is more important. The question clearly states that 10% data are missing.

upvoted 1 times

...

Stokvisss

10 months ago

Selected Answer: B

I'm going with B. 10% missing data on 10 years of data shouldn't matter too much, so D falls off. Seasonality introduces issues and should be fixed. A and C are wrong for obvious reasons.

upvoted 2 times

2eb8df0

3 months, 2 weeks ago

10% data in 10 years is an entire year of missing data. Its D, we dont even know if this company has seasonality problems

upvoted 1 times

...

aakash_0086

11 months, 1 week ago

Selected Answer: D

Linear interpolation would help to handle 10% missing values

upvoted 2 times

...

SVGoogle89

11 months, 1 week ago

B. 10% of the days data missing out of 365*10 days

upvoted 1 times

...

CloudHandsOn

11 months, 2 weeks ago

Selected Answer: D

D. Based on the problem, we need to address missing data and not seasonal variance.

upvoted 2 times

...

Oralinux

1 year ago

Answer: D

upvoted 1 times

...

aquanaveen

1 year ago

Selected Answer: B

B. Apply smoothing to correct for seasonal variation. Smoothing techniques, such as using moving averages or other time series smoothing methods, can help in reducing noise and capturing the underlying patterns in the sales data. Seasonal variation is a common issue in time series data, especially in retail where sales may exhibit regular patterns based on seasons, holidays, or other recurring events.

upvoted 3 times

...