Question 11
Content Domain 2: Exploratory Data AnalysisA machine learning team needs to assign categories to a large set of raw images before training a supervised model. Which tool is most appropriate for this task based on the source material?
Correct answer: B
Explanation
Data labelling tools are used to assign labels to raw data so it can be prepared for modeling. Amazon Mechanical Turk is an example of a data labelling tool. — Source material: Use data labelling tools (for example, Amazon Mechanical Turk).
Why each option is right or wrong
A. A data visualization dashboard for plotting model metrics
Data labelling tools assign labels to data before modeling, not display model performance charts.
B. Amazon Mechanical Turk for annotating the images
The source material identifies Amazon Mechanical Turk as an example of a data labelling tool. Because the team must assign categories to raw images before supervised training, a labeling tool that supports annotation is the appropriate choice here.
C. A feature scaling utility for normalizing numeric values
Feature scaling changes numeric ranges; it does not create labels for unlabeled images.
D. A model deployment service for publishing predictions
Deployment services serve trained models after preparation and training, not label source data.