AWS AWS Machine Learning and AI Services 4 — Questions and Answers
Question 1: Which AWS service provides a fully managed Jupyter notebook environment for data scientists to build and train ML models?
- Amazon SageMaker Studio (Correct answer)
- AWS Glue DataBrew
- Amazon EMR Notebooks
- AWS Lake Formation
Correct answer: Amazon SageMaker Studio
Amazon SageMaker Studio is a fully managed, web-based IDE that provides Jupyter notebooks along with tools for building, training, and deploying ML models.
Question 2: What does Amazon Rekognition Custom Labels allow you to do?
- Train a custom image classification model using your own labeled images (Correct answer)
- Detect faces in real-time video streams
- Translate text overlaid on images
- Generate synthetic training images using GANs
Correct answer: Train a custom image classification model using your own labeled images
Amazon Rekognition Custom Labels lets you train a custom computer vision model using your own labeled dataset without deep ML expertise.
Question 3: An ML team wants to detect data drift in a deployed SageMaker model. Which SageMaker feature should they enable?
- SageMaker Model Monitor (Correct answer)
- SageMaker Clarify
- SageMaker Debugger
- SageMaker Experiments
Correct answer: SageMaker Model Monitor
SageMaker Model Monitor continuously monitors deployed models for data quality drift, model quality drift, and bias drift.
Question 4: Which AWS AI service can automatically extract text, forms, and tables from scanned documents?
- Amazon Textract (Correct answer)
- Amazon Comprehend
- Amazon Kendra
- Amazon Rekognition
Correct answer: Amazon Textract
Amazon Textract uses ML to automatically extract printed text, handwriting, forms, and tables from documents without manual configuration.
Question 5: What is the purpose of SageMaker Feature Store?
- A centralized repository to store, share, and retrieve ML features for training and inference (Correct answer)
- A marketplace to purchase pre-built ML models
- A storage service for raw training datasets
- A feature flag system for A/B testing ML models
Correct answer: A centralized repository to store, share, and retrieve ML features for training and inference
SageMaker Feature Store provides a centralized store for ML features, enabling consistent feature definitions across training and real-time inference.
Question 6: Which Amazon Forecast algorithm is best suited when you have no historical data but have related time-series data?
- CNN-QR
- NPTS (Correct answer)
- DeepAR+
- Prophet
Correct answer: NPTS
NPTS (Non-Parametric Time Series) is designed for cases with sparse or no historical data and works with related time-series features.
Question 7: A company wants to add real-time personalized recommendations to their e-commerce website with minimal ML expertise required. Which AWS service is most appropriate?
- Amazon Personalize (Correct answer)
- Amazon SageMaker Autopilot
- AWS DeepComposer
- Amazon Forecast
Correct answer: Amazon Personalize
Amazon Personalize is a fully managed service that uses the same ML technology as Amazon.com to deliver real-time personalized recommendations.
AWS AWS Machine Learning and AI Services 4
Which AWS service provides a fully managed Jupyter notebook environment for data scientists to build and train ML models?