nerdexam
Amazon

AIF-C01 · Question #66

You are tasked with building a solution to classify customer reviews as positive or negative. The solution must handle real-time inferencing. Which of the following approaches and services are…

The correct answer is A. Use AWS Lambda for real-time predictions. C. Deploy the model using Amazon SageMaker Hosting Services. E. Use Amazon SageMaker to train the model. For a real-time sentiment classification solution, appropriate AWS services include Amazon SageMaker for model training and hosting, and AWS Lambda for serving real-time predictions.

Submitted by hans_de· Mar 30, 2026Applications of Foundation Models

Question

You are tasked with building a solution to classify customer reviews as positive or negative. The solution must handle real-time inferencing. Which of the following approaches and services are appropriate? (Choose three.)

Options

  • AUse AWS Lambda for real-time predictions.
  • BUse Amazon Polly for text-t o-speech capabilities.
  • CDeploy the model using Amazon SageMaker Hosting Services.
  • DUse AWS Glue for batch processing.
  • EUse Amazon SageMaker to train the model.

How the community answered

(31 responses)
  • A
    81% (25)
  • B
    6% (2)
  • D
    13% (4)

Why each option

For a real-time sentiment classification solution, appropriate AWS services include Amazon SageMaker for model training and hosting, and AWS Lambda for serving real-time predictions.

AUse AWS Lambda for real-time predictions.Correct

AWS Lambda can be used to invoke an endpoint for real-time predictions, acting as a serverless compute service to process incoming review requests and return classification results immediately.

BUse Amazon Polly for text-t o-speech capabilities.

Amazon Polly is a text-to-speech service, which is unrelated to classifying text reviews as positive or negative.

CDeploy the model using Amazon SageMaker Hosting Services.Correct

Amazon SageMaker Hosting Services provides robust, scalable endpoints for deploying trained models, which is crucial for handling real-time inference requests efficiently and reliably.

DUse AWS Glue for batch processing.

AWS Glue is an extract, transform, and load (ETL) service primarily used for batch processing and data preparation, not for real-time inferencing.

EUse Amazon SageMaker to train the model.Correct

Amazon SageMaker offers comprehensive capabilities for training machine learning models, including sentiment analysis models, by providing managed infrastructure and tools for data preparation, algorithm selection, and hyperparameter tuning.

Concept tested: Real-time ML inference architecture on AWS

Source: https://aws.amazon.com/sagemaker/features/model-hosting/

Topics

#real-time inference#SageMaker#sentiment classification#AWS Lambda

Community Discussion

No community discussion yet for this question.

Full AIF-C01 Practice