Using AI for the Optimization of Regression Test Suites CT-AI Practice Questions (Page 3)

Questions 11–15

1. 11foundation · easy

   An AI model predicts that only 20% of the regression tests are likely to detect faults in the new build. What is the most appropriate action based on this prediction?

   • ARun the other 80% of tests
   • BDelete the predicted 20% of tests
   • CRun all tests to be safe
   • DRun only the predicted 20% of tests
   Reveal Solution Discussion Select an answer first
2. 12foundation · easy

   Which of the following is a key benefit of applying AI to regression test suite optimization?

   • AIt ensures that every test case is executed in every regression run
   • BIt automatically fixes defects found by the regression suite
   • CIt removes the need for test maintenance
   • DIt reduces the time and cost of regression testing while maintaining coverage of critical risks
   Reveal Solution Discussion Select an answer first
4. 13expert · hard

   A team has a regression suite of 2,000 tests. They have trained an AI model to predict the probability of each test failing on a given code change. The model has a precision of 90% and a recall of 50%. They need to decide on a threshold for selecting tests. What is the most important trade-off to consider?

   • AThreshold does not affect the model's performance
   • BThreshold only affects execution time, not fault detection
   • CHigher threshold increases recall but decreases precision
   • DHigher threshold increases precision but decreases recall, potentially missing real failures
   Reveal Solution Discussion Select an answer first
5. 14foundation · easy

   An AI model assigns each test case a score based on the number of times it has failed in the past. Which prioritization factor does this model use?

   • AExecution time
   • BFailure history
   • CRequirement volatility
   • DCode coverage
   Reveal Solution Discussion Select an answer first
6. 15application · medium

   A QA team has a regression suite of 2,000 tests that takes 12 hours to run. They want to run a subset of tests on every code change, but they cannot afford to run the full suite. They have historical data on which tests have failed in the past and which code changes were associated with those failures. Which AI approach would best help them choose a subset of tests to run?

   • ATrain a machine learning model to predict the probability of each test failing on a given code change, then select tests above a threshold
   • BRun all tests that have ever failed in the past
   • CRun all tests that cover the most recently changed modules
   • DRandomly select 20% of the tests each time
   Reveal Solution Discussion Select an answer first