IBM A1000-108 - Assessment: Foundations of AI and Machine Learning Intermediate Practice Exam: Medium Difficulty 2025

A retail company wants to implement an AI system that can predict customer purchasing behavior based on browsing history, past purchases, and demographic data. The data science team needs to choose between supervised and unsupervised learning approaches. Which combination of learning type and algorithm would be most appropriate for this scenario?

During the data preparation phase for a machine learning project, a data scientist discovers that 15% of values in a critical feature column are missing. The feature shows strong correlation with the target variable. What is the most appropriate strategy to handle this missing data while preserving model performance?

A financial services company is developing an AI model to automate loan approval decisions. The model performs well overall but shows a pattern of denying loans to qualified applicants from certain demographic groups at a higher rate than others, despite similar credit profiles. Which aspect of responsible AI is being violated, and what should be the primary corrective action?

A healthcare organization is building a diagnostic AI system that needs to identify rare diseases from medical imaging. The training dataset contains 10,000 normal cases but only 100 cases of the rare disease. What combination of techniques would best address this class imbalance problem?

An AI system is being designed to power a chatbot for customer service. The system needs to understand customer intent, extract relevant entities from questions, and generate appropriate responses. Which combination of AI technologies would be most suitable for this multi-faceted application?

A data science team is preparing a dataset for training a machine learning model. They need to split the data into training, validation, and test sets. After splitting, they discover that the test set has a significantly different distribution of the target variable compared to the training set. What is the primary concern with this situation and how should it be addressed?

A manufacturing company wants to implement predictive maintenance using AI to predict equipment failures before they occur. They have 5 years of sensor data including temperature, vibration, pressure readings, and historical failure records. Which type of machine learning problem is this, and what key data preparation challenge must be addressed?

An organization is deploying an AI model that makes recommendations affecting people's employment opportunities. According to responsible AI principles, what governance practices should be implemented to ensure accountability and ongoing monitoring?

A data scientist is evaluating two machine learning models for a fraud detection system. Model A has 95% accuracy, while Model B has 88% accuracy. However, Model B has significantly better precision and recall for the fraud class (minority class). The dataset contains 5% fraudulent transactions and 95% legitimate transactions. Which model should be selected and why?

A company is implementing an AI system that will process customer data to provide personalized marketing recommendations. The system will use personal information including browsing history, purchase patterns, and demographic data. Which combination of responsible AI practices should be prioritized to address both privacy and transparency concerns?