AWS Certified Machine Learning – Specialty — Question 217
A data scientist at a retail company is forecasting sales for a product over the next 3 months. After preliminary analysis, the data scientist identifies that sales are seasonal and that holidays affect sales. The data scientist also determines that sales of the product are correlated with sales of other products in the same category.
The data scientist needs to train a sales forecasting model that incorporates this information.
Which solution will meet this requirement with the LEAST development effort?
Answer options
- A. Use Amazon Forecast with Holidays featurization and the built-in autoregressive integrated moving average (ARIMA) algorithm to train the model.
- B. Use Amazon Forecast with Holidays featurization and the built-in DeepAR+ algorithm to train the model.
- C. Use Amazon SageMaker Processing to enrich the data with holiday information. Train the model by using the SageMaker DeepAR built-in algorithm.
- D. Use Amazon SageMaker Processing to enrich the data with holiday information. Train the model by using the Gluon Time Series (GluonTS) toolkit.
Correct answer: B
Explanation
Option B is the correct answer because Amazon Forecast with the built-in DeepAR+ algorithm is specifically designed for time series forecasting and handles seasonality and related products effectively with minimal configuration. Options A, C, and D involve either more complex configurations or different tools that require additional development effort compared to using the built-in capabilities of Amazon Forecast.