MLA-C01 Question 76
Select 3You are building a machine learning model to classify product images into various categories for an e-commerce platform. The success of your model depends on a high-quality labeled dataset. You decide to use Amazon SageMaker Ground Truth to assist with data labeling. Which features of SageMaker Ground Truth can help you ensure high-quality labeled data?
- A
Automated data labeling using machine learning models to reduce labeling costs
- B
Integration with human labelers and managed workforce for complex labeling tasks
- C
Direct deployment of the labeled dataset into production without model validation
- D
Annotation consolidation to improve labeling quality using multiple annotators
- E
Support for only image-based labeling tasks
Show answer and explanation
Correct answers: A, B, D
Explanation
Amazon SageMaker Ground Truth is a powerful service for creating high-quality labeled datasets. It achieves this through automated data labeling, integration with human labelers for complex tasks, and annotation consolidation to resolve disagreements and improve label quality. These features make it highly suitable for building datasets for machine learning models, such as the product image classification model in this scenario.
- A. Correct.
Amazon SageMaker Ground Truth provides automated data labeling, which uses machine learning models to pre-label data and reduces the cost and time required for manual labeling.
- B. Correct.
SageMaker Ground Truth integrates with human labelers and managed workforces, such as Amazon Mechanical Turk or third-party vendors, to handle complex or edge-case labeling tasks.
- C. Incorrect.
Direct deployment of the labeled dataset without validation is not a feature of SageMaker Ground Truth. Labeled datasets should always be tested and validated before production use.
- D. Correct.
Annotation consolidation in SageMaker Ground Truth ensures higher-quality labels by using multiple annotators for the same data point and resolving disagreements through advanced algorithms.
- E. Incorrect.
SageMaker Ground Truth supports multiple data types, including image, text, video, and 3D point cloud data, not just image-based tasks.