Amazon AWS Certified AI Practitioner AIF-C01 Practice Test 7
Question 60
The initial scenario was: An airline needs a conversational chatbot that recognizes traveler intent and manages slot values during booking. The original control is already in place. Now the company later wants to convert agent response text into voice. Which ADDITIONAL action BEST addresses the new condition?
- Pair the dialog flow with Amazon Polly speech synthesis
- Test utterances map to booking intents and collect required travel slots
- Use Amazon Lex for intent-based conversational interfaces
- Use Amazon S3 alone to interpret booking dialogue
Explanation
Correct Answer
A:The company later wants to convert agent response text into voice. The appropriate added action is: Pair the dialog flow with Amazon Polly speech synthesis.
Incorrect Answers
B:This describes evidence for the original approach, not a new action: Test utterances map to booking intents and collect required travel slots.
C:This was the original control and is already implemented. Lex supports conversational intent recognition and slot collection.
D:This original shortcut remains unsuitable: Object storage does not supply natural-language intent recognition.
Question 61
A warehouse wants to identify products and read visual clues in photographs for inventory checks. Which TWO statements jointly identify the best initial approach and an observable verification result? (Select TWO.)
- Select computer vision image analysis for photographed inventory
- Use a time-series forecasting model as the only photo detector
- Use forecasting on historical sales time series
- Use a recommendation approach informed by interaction history
- Held-out inventory photos yield appropriate visual detections for review
Explanation
Correct Answers
A:Correct approach. Visual object detection and image processing operate on image content. This directly serves the stated situation: A warehouse wants to identify products and read visual clues in photographs for inventory checks.
E:Correct operating evidence. Observing held-out inventory photos yield appropriate visual detections for review verifies this approach instead of relying on unperformed plans.
Incorrect Answers
B:Forecasting future numerical sequences does not inspect image pixels. The suggested shortcut would not meet the present requirement.
C:This would be appropriate if the separate requirement arose: Managers later seek the next month's demand by SKU. That is not the present condition.
D:This addresses a different scenario: A marketing analyst has purchase history and wants to suggest products based on customer affinities. It does not constitute the requested approach or operating result for the current case.
Question 62
A warehouse wants to identify products and read visual clues in photographs for inventory checks. A colleague suggests: Use a time-series forecasting model as the only photo detector. Which finding BEST explains why the suggestion should be rejected?
- Forecasting future numerical sequences does not inspect image pixels
- A generative prediction is unsuitable for fixed legal arithmetic and could hallucinate
- The shortcut is safe because its terminology is familiar to the team
- Alphabetic order does not model individual preference signals
Explanation
Correct Answer
A:Forecasting future numerical sequences does not inspect image pixels. This is the direct defect in the proposed action: Use a time-series forecasting model as the only photo detector.
Incorrect Answers
B:This concern belongs to another situation: A tax calculation uses a fixed statutory bracket and must always return the same exact result for the same numbers.
C:Familiar wording is not proof that an action is appropriate. The actual defect is: Forecasting future numerical sequences does not inspect image pixels.
D:This finding is associated with a different problem: A marketing analyst has purchase history and wants to suggest products based on customer affinities.
Question 63
A warehouse wants to identify products and read visual clues in photographs for inventory checks. After implementing Select computer vision image analysis for photographed inventory, which observation BEST verifies the intended result?
- The team approves Select computer vision image analysis for photographed inventory but does not run an operational test
- The team schedules a future test of whether held-out inventory photos yield appropriate visual detections for review, but has no result
- Held-out inventory photos yield appropriate visual detections for review
- Offline ranking and controlled experiments show more relevant recommendations
Explanation
Correct Answer
C:The observed result, Held-out inventory photos yield appropriate visual detections for review, provides direct evidence. Visual object detection and image processing operate on image content.
Incorrect Answers
A:The required observable condition is: Held-out inventory photos yield appropriate visual detections for review.
B:Verify the actual outcome: Held-out inventory photos yield appropriate visual detections for review.
D:This observation would be evidence for another case: A marketing analyst has purchase history and wants to suggest products based on customer affinities, not the current scenario.
Question 64
The initial scenario was: A warehouse wants to identify products and read visual clues in photographs for inventory checks. The original control is already in place. Now managers later seek the next month's demand by sku. Which ADDITIONAL action BEST addresses the new condition?
- Use forecasting on historical sales time series
- Held-out inventory photos yield appropriate visual detections for review
- Select computer vision image analysis for photographed inventory
- Use a time-series forecasting model as the only photo detector
Explanation
Correct Answer
A:Managers later seek the next month's demand by SKU. The appropriate added action is: Use forecasting on historical sales time series.
Incorrect Answers
B:This describes evidence for the original approach, not a new action: Held-out inventory photos yield appropriate visual detections for review.
C:This was the original control and is already implemented. Visual object detection and image processing operate on image content.
D:This original shortcut remains unsuitable: Forecasting future numerical sequences does not inspect image pixels.
Question 65
A marketing analyst has purchase history and wants to suggest products based on customer affinities. Which option BEST meets the stated AI/ML or AWS requirement?
- Implement deterministic rules instead of probabilistic ML for this statutory calculation
- Apply fraud or anomaly detection to transaction behavior
- Use deterministic alphabetical sorting as a substitute for personalization
- Use a recommendation approach informed by interaction history
Explanation
Correct Answer
D:Recommendations rank likely relevant items using customer behavior signals. In this situation, choose: Use a recommendation approach informed by interaction history.
Incorrect Answers
A:This can be valid for a different task: A tax calculation uses a fixed statutory bracket and must always return the same exact result for the same numbers. The current request is different and needs the selected approach.
B:This is appropriate only after a distinct change: The store later needs to detect suspicious checkout patterns. It is not the best initial response to a marketing analyst has purchase history and wants to suggest products based on customer affinities.
C:Alphabetic order does not model individual preference signals. The proposal, Use deterministic alphabetical sorting as a substitute for personalization, does not satisfy this need.
Question 66
A marketing analyst has purchase history and wants to suggest products based on customer affinities. A colleague suggests: Use deterministic alphabetical sorting as a substitute for personalization. Which finding BEST explains why the suggestion should be rejected?
- Image creation neither models transaction labels nor meets the required decision timing
- Alphabetic order does not model individual preference signals
- The shortcut is safe because its terminology is familiar to the team
- A generative prediction is unsuitable for fixed legal arithmetic and could hallucinate
Explanation
Correct Answer
B:Alphabetic order does not model individual preference signals. This is the direct defect in the proposed action: Use deterministic alphabetical sorting as a substitute for personalization.
Incorrect Answers
A:This concern belongs to another situation: A finance firm has labeled historical credit-card transactions and wants real-time alerts on likely fraudulent purchases.
C:Familiar wording is not proof that an action is appropriate. The actual defect is: Alphabetic order does not model individual preference signals.
D:This finding is associated with a different problem: A tax calculation uses a fixed statutory bracket and must always return the same exact result for the same numbers.
Question 67
A marketing analyst has purchase history and wants to suggest products based on customer affinities. After implementing Use a recommendation approach informed by interaction history, which observation BEST verifies the intended result?
- Test inputs match the published statutory formula exactly
- The team schedules a future test of whether offline ranking and controlled experiments show more relevant recommendations, but has no result
- The team approves Use a recommendation approach informed by interaction history but does not run an operational test
- Offline ranking and controlled experiments show more relevant recommendations
Explanation
Correct Answer
D:The observed result, Offline ranking and controlled experiments show more relevant recommendations, provides direct evidence. Recommendations rank likely relevant items using customer behavior signals.
Incorrect Answers
A:This observation would be evidence for another case: A tax calculation uses a fixed statutory bracket and must always return the same exact result for the same numbers, not the current scenario.
B:Verify the actual outcome: Offline ranking and controlled experiments show more relevant recommendations.
C:The required observable condition is: Offline ranking and controlled experiments show more relevant recommendations.
Question 68
The initial scenario was: A marketing analyst has purchase history and wants to suggest products based on customer affinities. The original control is already in place. Now the store later needs to detect suspicious checkout patterns. Which ADDITIONAL action BEST addresses the new condition?
- Use deterministic alphabetical sorting as a substitute for personalization
- Apply fraud or anomaly detection to transaction behavior
- Offline ranking and controlled experiments show more relevant recommendations
- Use a recommendation approach informed by interaction history
Explanation
Correct Answer
B:The store later needs to detect suspicious checkout patterns. The appropriate added action is: Apply fraud or anomaly detection to transaction behavior.
Incorrect Answers
A:This original shortcut remains unsuitable: Alphabetic order does not model individual preference signals.
C:This describes evidence for the original approach, not a new action: Offline ranking and controlled experiments show more relevant recommendations.
D:This was the original control and is already implemented. Recommendations rank likely relevant items using customer behavior signals.
Question 69
A tax calculation uses a fixed statutory bracket and must always return the same exact result for the same numbers. Which option BEST meets the stated AI/ML or AWS requirement?
- Evaluate an ML risk classification model with labeled review outcomes
- Train a generative model to invent tax brackets from examples
- Use supervised fraud classification with fraud labels and a real-time decision path
- Implement deterministic rules instead of probabilistic ML for this statutory calculation
Explanation
Correct Answer
D:Known precise arithmetic rules are more appropriate than uncertain predictions. In this situation, choose: Implement deterministic rules instead of probabilistic ML for this statutory calculation.
Incorrect Answers
A:This is appropriate only after a distinct change: The team next wants to predict which filings may need manual review. It is not the best initial response to a tax calculation uses a fixed statutory bracket and must always return the same exact result for the same numbers.
B:A generative prediction is unsuitable for fixed legal arithmetic and could hallucinate. The proposal, Train a generative model to invent tax brackets from examples, does not satisfy this need.
C:This can be valid for a different task: A finance firm has labeled historical credit-card transactions and wants real-time alerts on likely fraudulent purchases. The current request is different and needs the selected approach.
Question 70
A tax calculation uses a fixed statutory bracket and must always return the same exact result for the same numbers. A colleague suggests: Train a generative model to invent tax brackets from examples. Which finding BEST explains why the suggestion should be rejected?
- A generative prediction is unsuitable for fixed legal arithmetic and could hallucinate
- Image creation neither models transaction labels nor meets the required decision timing
- Polly generates speech from text rather than transcribing audio
- The shortcut is safe because its terminology is familiar to the team
Explanation
Correct Answer
A:A generative prediction is unsuitable for fixed legal arithmetic and could hallucinate. This is the direct defect in the proposed action: Train a generative model to invent tax brackets from examples.
Incorrect Answers
B:This finding is associated with a different problem: A finance firm has labeled historical credit-card transactions and wants real-time alerts on likely fraudulent purchases.
C:This concern belongs to another situation: A call center stores English phone recordings and needs searchable written transcripts of the speech.
D:The actual defect is: A generative prediction is unsuitable for fixed legal arithmetic and could hallucinate.