Translate Business Use Case Goals into a Description of the Desired Inputs and Outputs for the AI Pipeline GENERATIVE-AI-ENGINEER-ASSOCIATE Practice Questions (Page 3)

Questions 11–15

1. 11application · medium

   A logistics company wants to build a pipeline that predicts delivery delays. They have historical shipment data with timestamps, routes, and weather conditions. The operations team needs a daily report of shipments likely to be delayed by more than 2 hours. What input specification is most appropriate for this pipeline?

   • ACustomer feedback surveys about past deliveries.
   • BReal-time streaming of GPS coordinates for every shipment.
   • CHistorical data only, without any new inputs.
   • DDaily batch of shipment records with route, scheduled delivery time, and forecasted weather for the next 24 hours.
   Reveal Solution Discussion Select an answer first
2. 12expert · hard

   A hospital wants to build an AI pipeline to predict patient readmission risk within 30 days of discharge. The goal is to reduce readmissions by 15%. They have electronic health records (EHR), lab results, and medication data. The care team needs a list of high-risk patients at discharge to schedule follow-up appointments. However, the hospital's data is fragmented across legacy systems, and the pipeline must be operational within 3 months. What is the most critical factor to address when defining the pipeline's inputs and outputs?

   • ARequire the pipeline to output a detailed clinical report for each patient.
   • BUse only the most recent EHR data, ignoring lab results and medication data.
   • CEnsure the input data from all legacy systems can be integrated and standardized into a unified format for the model.
   • DDefine the output as a list of patients with a binary high-risk flag, without probabilities.
   Reveal Solution Discussion Select an answer first
4. 13foundation · easy

   For a sentiment analysis pipeline that processes product reviews, which input specification is necessary?

   • AThe review text, the product ID, and the review date, with text normalized to lowercase.
   • BA list of positive and negative words.
   • CThe sentiment label for each review.
   • DThe number of reviews to process.
   Reveal Solution Discussion Select an answer first
5. 14application · medium

   A telecom company wants to use AI to reduce customer service call volume. They have call transcripts, customer account data, and product usage data. The goal is to identify customers who are likely to call with billing issues. What is the most appropriate measurable objective?

   • AClassify all calls into categories such as billing, technical, or sales.
   • BProvide a dashboard of call volume trends by issue type.
   • CReduce call volume by 10% within 6 months.
   • DGenerate a daily list of customers with a predicted probability of calling about a billing issue, with a target precision of at least 80%.
   Reveal Solution Discussion Select an answer first
6. 15foundation · easy

   A healthcare provider wants to reduce patient no-shows. Which objective best translates this goal into a clear purpose for an AI pipeline?

   • AAnalyze appointment data to find patterns.
   • BReduce the number of missed appointments.
   • CSend reminders to all patients before their appointments.
   • DPredict the probability of no-show for each scheduled appointment, with the goal of flagging high-risk appointments for reminder interventions.
   Reveal Solution Discussion Select an answer first