Using AI to Analyze Reported Defects CT-AI Practice Questions (Page 5)

Questions 21–23

1. 21application · medium

   A software company wants to proactively identify which modules are most likely to have defects in the next release, based on historical defect data, code complexity metrics, and recent change frequency. Which approach should they take?

   • AApply NLP to analyze defect descriptions and count the most frequent error messages.
   • BManually review all historical defects and assign risk scores based on personal experience.
   • CUse clustering to group modules with similar defect histories and assume the largest cluster is highest risk.
   • DTrain a supervised classification model on historical data to predict high-risk modules for the upcoming release.
   Reveal Solution Discussion Select an answer first
2. 22foundation · easy

   In AI-assisted defect triage, what does the AI typically use to prioritize defects?

   • AThe order in which defects were reported
   • BThe name of the tester who reported the defect
   • CThe length of the defect description
   • DSeverity, impact, and likelihood of occurrence
   Reveal Solution Discussion Select an answer first
3. 23application · medium

   A development team wants to use AI to predict which areas of the application are likely to have the most defects in the next sprint, based on the number of code changes and past defect density. They also want to automatically create work items for the predicted high-risk areas. What should they do?

   • AUse a rule-based system that creates a work item for every module with more than 10 code changes.
   • BUse NLP to analyze defect descriptions and create work items for defects with the word 'critical'.
   • CUse a clustering algorithm to group code changes and manually create work items for the largest cluster.
   • DTrain a predictive model on historical data and integrate its output with the project management tool to create work items.
   Reveal Solution Discussion Select an answer first