IBM A1000-050 - Assessment: Foundations of AI Intermediate Practice Exam: Medium Difficulty 2025

A healthcare organization is developing an AI system to assist radiologists in detecting anomalies in X-ray images. The system needs to learn from a labeled dataset of 50,000 images where each image is tagged as either 'normal' or 'abnormal'. Which combination of AI approach and learning paradigm is most appropriate for this scenario?

A retail company wants to segment its customer base into distinct groups based on purchasing behavior, browsing patterns, and demographics without predefined categories. The marketing team wants to discover natural groupings in the data to create targeted campaigns. Which machine learning approach and specific technique would be most effective?

An AI development team notices that their fraud detection model performs exceptionally well on training data (99% accuracy) but poorly on new, real-world transactions (65% accuracy). The model flags many legitimate transactions as fraudulent. What is the primary issue, and what is the best approach to address it?

A financial services company is implementing an AI system to automatically approve or deny loan applications. Regulatory requirements mandate that the company must be able to explain to customers why their application was denied. Which AI model characteristic and approach should be prioritized?

A manufacturing company wants to implement an AI system to optimize robot movements in real-time on an assembly line. The robots need to learn the most efficient paths through trial and error, receiving positive feedback for faster assembly times and negative feedback for errors. Which learning paradigm is most appropriate?

A global e-commerce platform is deploying an AI-powered recommendation system that will influence product visibility and sales. During testing, the team discovers the system consistently recommends higher-priced items to users from certain demographic groups. What combination of AI ethics principles and actions should the team prioritize?

A telecommunications company wants to predict customer churn by analyzing call records, data usage, customer service interactions, and billing history. They have historical data for 2 million customers with labels indicating whether each customer left the service. After model training, what metric combination would best evaluate the model's performance given that only 5% of customers typically churn?

A logistics company is implementing an AI solution to optimize delivery routes across multiple cities. The system needs to process real-time traffic data, weather conditions, and delivery priorities while considering fuel costs and delivery time windows. Which AI application category and implementation approach best fits this scenario?

An insurance company is developing an AI system to process and extract information from various document types including handwritten claim forms, printed policies, and scanned medical records. The system needs to understand document structure, extract key data fields, and categorize information. Which combination of AI technologies would be most effective?

A healthcare AI research team is developing a diagnostic model using patient data from multiple hospitals. They need to ensure patient privacy while still benefiting from the collective data. Hospital policies prevent sharing raw patient data between institutions. Which AI approach and privacy-preserving technique would allow collaborative model training?