A data science team is planning to build a natural language processing (NLP) application. The application's text preprocessing stage will include part-of-speech tagging and key phase extraction. The…

Question

Options

• AUse Amazon Comprehend for the part-of-speech tagging, key phase extraction, and classification
• BUse an NLP library in Amazon SageMaker for the part-of-speech tagging.
• CUse Amazon Comprehend for the part-of-speech tagging and key phase extraction tasks.
• DUse Amazon Comprehend for the part-of-speech tagging and key phase extraction tasks.

• A
  84% (21)
• B
  4% (1)
• C
  8% (2)
• D
  4% (1)

Explanation

Amazon Comprehend natively supports both part-of-speech (POS) tagging and key phrase extraction as built-in API operations requiring no model training or configuration. Using Comprehend for these preprocessing steps eliminates the need to build, train, or deploy any custom NLP preprocessing components, which is the fastest possible approach. The custom MXNet classification model can then be hosted on Amazon SageMaker to handle the final classification step. Option A represents the fastest end-to-end build because it maximizes use of managed services for the preprocessing pipeline while still accommodating the team's existing custom model. Options C and D appear similar but may differ in how the MXNet model is deployed or integrated. Option B suggests using an unmanaged NLP library within SageMaker for preprocessing, which requires more setup than simply calling Comprehend APIs. The principle here is: use fully managed services wherever possible to minimize build time.

No community discussion yet for this question.