Identify Benefits and Capabilities of Generative AI Solutions AB-731 Practice Questions (Page 2)

Questions 6–10

1. 6expert · hard

   A healthcare provider wants to build a machine learning model to predict patient readmission risk. They have electronic health records (EHR) data, but the data is incomplete and contains many missing values. The provider wants to deploy the model in a clinical setting where false negatives (missing a high-risk patient) are dangerous. What is the most important action to take?

   • AUse a generative AI model to fill in the missing values.
   • BExclude all records with missing values to ensure data quality.
   • CUse a model that is robust to missing data, such as a tree-based model, and evaluate its performance with a focus on recall.
   • DImpute missing values using a simple method like mean imputation.
   Reveal Solution Discussion Select an answer first
2. 7application · medium

   A company is deploying an AI-powered document summarization service that processes confidential legal documents. The service is accessed by employees through a web application. Which security consideration is most critical to address?

   • AStore the documents in an encrypted database.
   • BImplement authentication and authorization so only authorized employees can submit documents and view summaries.
   • CEnsure the web application uses HTTPS to encrypt data in transit.
   • DUse a private endpoint for the AI service to avoid public internet exposure.
   Reveal Solution Discussion Select an answer first
4. 8foundation · easy

   Which scenario is most suitable for applying machine learning to provide significant business value?

   • ASending a confirmation email after a purchase
   • BCalculating the total sales for a quarter using a spreadsheet formula
   • CStoring customer records in a relational database
   • DPredicting customer churn based on historical usage patterns
   Reveal Solution Discussion Select an answer first
5. 9application · medium

   A bank wants to use machine learning to approve loan applications. They have historical data on approved and rejected loans. The data includes applicant demographics, income, and credit history. What is a critical data quality issue that could affect the model's fairness and performance?

   • AThe data is stored in multiple formats.
   • BThe data is not labeled with the correct loan amounts.
   • CThe data may contain biases, such as underrepresentation of certain demographic groups.
   • DThe data is too large to process quickly.
   Reveal Solution Discussion Select an answer first
6. 10expert · hard

   A customer service organization uses a generative AI system to respond to customer complaints. The system sometimes generates responses that are too defensive or argumentative, damaging customer satisfaction. The team has tried adjusting the prompt, but the issue persists. They have a limited budget and need a solution that improves the tone without retraining the model. What should they do?

   • AAdd more disclaimers to the responses to reduce defensiveness.
   • BUse a few-shot prompting technique with examples of empathetic and effective responses.
   • CUse a different language model that is known for being more empathetic.
   • DIncrease the model's temperature to make responses more varied.
   Reveal Solution Discussion Select an answer first