AWS Certified Solutions Architect - Associate
Get started today
Ultimate access to all questions.
Deep dive into the quiz with AI chat providers.
We prepare a focused prompt with your quiz and certificate details so each AI can offer a more tailored, in-depth explanation.
A company is running a popular social media website. The website gives users the ability to upload images to share with other users. The company wants to make sure that the images do not contain inappropriate content. The company needs a solution that minimizes development effort.
What should a solutions architect do to meet these requirements?
Other
Community
UAnonymous
A
Use Amazon Comprehend to detect inappropriate content. Use human review for low-confidence predictions.
B
Use Amazon Rekognition to detect inappropriate content. Use human review for low-confidence predictions.
C
Use Amazon SageMaker to detect inappropriate content. Use ground truth to label low-confidence predictions.
D
Use AWS Fargate to deploy a custom machine learning model to detect inappropriate content. Use ground truth to label low-confidence predictions.
Explanation:
Explanation
Amazon Rekognition is the correct choice because:
-
Image-specific service: Amazon Rekognition is specifically designed for image and video analysis, including content moderation features that can detect inappropriate content in images.
-
Minimizes development effort: Amazon Rekognition is a fully managed service that requires minimal development effort compared to building custom models. The company can simply call the Rekognition API to analyze uploaded images.
-
Built-in content moderation: Rekognition has pre-trained models for detecting inappropriate content, including explicit nudity, suggestive content, violence, and other categories.
-
Human review integration: The solution can be designed to use human review for low-confidence predictions, which is a common pattern in content moderation workflows.
Why other options are incorrect:
-
A. Amazon Comprehend: This service is for natural language processing (text analysis), not image analysis. It cannot analyze image content.
-
C. Amazon SageMaker: While SageMaker can be used to build custom ML models for image analysis, this requires significant development effort to collect data, train models, and deploy them - which contradicts the requirement to minimize development effort.
-
D. AWS Fargate with custom ML model: This requires building and deploying a custom machine learning model, which involves substantial development effort for data collection, model training, deployment, and maintenance.
Best Practice: For content moderation of images with minimal development effort, Amazon Rekognition's content moderation feature is the most appropriate AWS service. It provides pre-trained models that can be easily integrated via API calls, and the confidence scores returned can be used to route low-confidence predictions to human reviewers.
Comments
Loading comments...