Hands-On Exercise: Exploratory Testing and Exploratory Data Analysis (EDA) CT-AI Practice Questions (Page 3)

Questions 11–15

1. 11foundation · easy

   Which of the following activities is a typical part of exploratory data analysis (EDA) for an AI dataset?

   • ADeploying the trained model to a production environment.
   • BCalculating summary statistics and plotting distributions of features.
   • CWriting unit tests for the AI model's code.
   • DSetting up a continuous integration pipeline for model retraining.
   Reveal Solution Discussion Select an answer first
2. 12foundation · easy

   What is the primary characteristic of exploratory testing that distinguishes it from scripted testing?

   • AIt relies on predefined test cases with expected results.
   • BIt is performed only after all scripted tests have passed.
   • CIt involves simultaneous learning, test design, and test execution.
   • DIt is automated and requires no human judgment.
   Reveal Solution Discussion Select an answer first
4. 13expert · hard

   A team is testing an AI system that predicts patient readmission risk. Exploratory testing reveals that the model performs poorly for patients over 80 years old. EDA shows that the training data has very few examples of patients in this age group. The team must decide how to proceed. Which action best integrates the findings from both activities?

   • AAdjust the model's decision threshold to be more conservative for all patients to reduce overall risk.
   • BRemove all patients over 80 from the test set to avoid penalizing the model for a demographic it was not trained on.
   • CRetrain the model using a more complex algorithm that can generalize better from limited data.
   • DPrioritize collecting more training data for patients over 80, and use the exploratory testing findings to define specific test cases for this demographic.
   Reveal Solution Discussion Select an answer first
5. 14application · medium

   A data analyst is performing EDA on a dataset for a churn-prediction model. The analyst creates a histogram of the 'customer_tenure' feature and notices a large spike at the value 0. What is the most appropriate next step?

   • AReplace tenure = 0 with the median tenure value.
   • BInvestigate the rows with tenure = 0 to understand if they represent new customers or a data recording error.
   • CRemove all rows with tenure = 0 to avoid skewing the model.
   • DIgnore the spike because histograms often have artifacts.
   Reveal Solution Discussion Select an answer first
6. 15application · medium

   A test lead is introducing exploratory testing to a team that is used to scripted testing. A team member asks, 'How is exploratory testing different from scripted testing?' What is the most accurate distinction to explain?

   • AExploratory testing does not require documentation, while scripted testing requires detailed test cases.
   • BIn exploratory testing, test design and test execution happen simultaneously, whereas in scripted testing, tests are designed in advance.
   • CExploratory testing is performed by testers, while scripted testing is performed by developers.
   • DExploratory testing is only used for AI systems, while scripted testing is used for traditional software.
   Reveal Solution Discussion Select an answer first