IBM A1000-076 - Assessment: Foundations of AI Intermediate Practice Exam: Medium Difficulty 2025

A retail company wants to implement an AI system that can predict customer churn based on historical purchase data, browsing behavior, and customer service interactions. The data scientist needs to choose between supervised and unsupervised learning approaches. Which approach is most appropriate and why?

An organization is developing a conversational AI assistant using IBM Watson. The assistant needs to understand user intents, extract relevant entities, and maintain context across multiple conversation turns. Which combination of Watson services would best address these requirements?

A healthcare provider is implementing an AI system to assist radiologists in detecting anomalies in medical images. During testing, the system shows 95% accuracy but has a high number of false negatives (missing actual diseases). What is the primary concern from an AI ethics perspective, and what metric should be prioritized?

A data science team is building a neural network for image classification. After training, the model performs exceptionally well on the training data (98% accuracy) but poorly on the validation data (72% accuracy). What problem is the model experiencing and what are the most effective remediation strategies?

An e-commerce company wants to use IBM Watson to analyze customer reviews and feedback to understand sentiment and extract key themes about product features. The data includes both structured ratings and unstructured text comments in multiple languages. Which Watson service combination would be most effective?

A financial institution is developing an AI model to automate loan approval decisions. Regulators require that the institution be able to explain why any loan application was denied. Which AI concept is most critical for meeting this regulatory requirement, and what approach should be implemented?

A manufacturing company is implementing a predictive maintenance system using sensor data from machinery. The system needs to identify when equipment is likely to fail within the next 48 hours. The data scientist must select appropriate features from thousands of sensor readings. What is this process called and what is a key consideration?

A global corporation needs to process and understand large volumes of internal documents, contracts, and reports to quickly find relevant information and insights. The solution must handle various document formats and understand document structure and relationships between concepts. Which IBM Watson service is most appropriate?

A technology company is deploying a facial recognition system for building access control. During testing, they discover the system has significantly lower accuracy for certain demographic groups. What type of AI problem is this, and what is the most appropriate first step to address it?

A data scientist is training a deep learning model for natural language processing and notices that the loss decreases during early epochs but then starts to increase on the validation set while continuing to decrease on the training set. The learning rate is set to 0.01. What combination of adjustments would most effectively address this issue?