Describe the Impact of Predictive AI CERTIFIED-AI-EXPERT Practice Questions (Page 1)

Questions 1–5

1. 1application · medium

   A bank wants to use predictive AI to reduce loan default risk. They have historical loan application data with outcomes. Which approach best leverages predictive AI to achieve this business outcome?

   • ATrain a model to predict the probability of default for each new applicant and use it to adjust credit decisions
   • BUse a generative model to create synthetic loan agreements for new customers
   • CDeploy a chatbot to answer customer questions about loan terms
   • DImplement a rule-based system that rejects all applicants with a credit score below 600
   Reveal Solution Discussion Select an answer first
2. 2expert · hard

   A hospital uses a predictive model to identify patients at high risk of sepsis. The model was trained on data from a single region and now shows lower accuracy for patients from other regions. The hospital must maintain high accuracy across all regions while ensuring the model's decisions can be explained to clinicians. What is the best approach?

   • AFreeze the current model and manually adjust predictions for other regions
   • BUse a complex deep learning model to capture all regional variations
   • CDeploy separate models for each region without sharing data
   • DRetrain the model with diverse data from all regions and use interpretable models where possible
   Reveal Solution Discussion Select an answer first
4. 3foundation · easy

   What is the first stage in the lifecycle of a predictive AI model?

   • AHyperparameter tuning
   • BData collection
   • CModel deployment
   • DModel monitoring
   Reveal Solution Discussion Select an answer first
5. 4expert · hard

   A logistics company wants to implement predictive AI to optimize delivery routes. They have historical delivery data, but the data is stored in siloed systems across different regions. The company also needs to ensure the model remains accurate as traffic patterns change. What is the best strategy?

   • AUse a generative model to create synthetic delivery data for missing regions
   • BIntegrate the data into a central repository, train a model, and set up continuous monitoring for drift
   • CTrain separate models for each region and combine their predictions
   • DDeploy the model immediately with the available data and adjust later
   Reveal Solution Discussion Select an answer first
6. 5application · medium

   A retail chain wants to reduce inventory costs by predicting which products will sell out quickly. They have historical sales data, but it is incomplete for some products. What is the best way to use predictive AI to achieve this goal?

   • AIgnore products with incomplete data and only forecast for complete ones
   • BManually review sales data for each product to predict demand
   • CUse a generative model to create sales data for missing products
   • DTrain a model on the available sales data to forecast demand for each product
   Reveal Solution Discussion Select an answer first