AI-Based System Testing Fundamentals CT-AI Practice Questions (Page 4)

Questions 16–20

1. 16application · medium

   A team is testing an AI model that classifies emails as spam or not spam. The model was trained on a dataset where 95% of emails are not spam. During testing, the model correctly classifies 90% of non-spam emails but only 50% of spam emails. What does this indicate?

   • AThe model is overfitting the training data.
   • BThe model is performing well because it correctly classifies most emails.
   • CThe model is biased towards the majority class (non-spam).
   • DThe test data is not representative of the training data.
   Reveal Solution Discussion Select an answer first
2. 17application · easy

   A company is developing an AI system to recommend personalized content to users. The specification states that the system should increase user engagement by 10% within six months. Which component of the AI system specification does this requirement belong to?

   • APerformance requirements
   • BInput data testing requirements
   • CModel requirements
   • DData requirements
   Reveal Solution Discussion Select an answer first
4. 18application · easy

   An AI system for predicting equipment failure in a factory uses sensor data. The test data contains many duplicate records due to a logging error. What is the most appropriate action to take?

   • AIgnore the duplicates because they are a minor issue.
   • BRemove duplicate records from the test data to ensure each event is counted once.
   • CReplace duplicates with synthetic data to increase the test set size.
   • DKeep the duplicates because they reflect real-world data collection issues.
   Reveal Solution Discussion Select an answer first
5. 19application · medium

   A team is testing a machine learning model for predicting house prices. The model performs well on the training data but poorly on the test data. What is the most likely issue?

   • AThe model is underfitting the training data.
   • BThe model is overfitting the training data.
   • CThe test data is not labeled correctly.
   • DThe test data is too small.
   Reveal Solution Discussion Select an answer first
6. 20foundation · easy

   Which testing activity is specifically focused on verifying that input data for an AI system is complete and within expected ranges?

   • AData quality testing
   • BSecurity testing
   • CPerformance testing
   • DModel accuracy testing
   Reveal Solution Discussion Select an answer first