Data, Models, and Behavioral Insights

This unit examines data requirements, data augmentation strategies, and the landscape of AI model families to turn behavioral signals into actionable insights. Students learn to design experiments, choose minimum viable datasets, compare classical/ML/deep/generative approaches, and apply advanced segmentation and intent-prediction methods to map customer journeys.

Learning Objectives

• Analyze trade-offs among data volume, data quality, and the Minimum Viable Dataset (MVD) to determine realistic data needs for behavioral AI experiments
• Apply synthetic data generation and survey-data techniques to augment limited datasets and validate resulting datasets using appropriate evaluation metrics
• Demonstrate experimental designs that compare classical AI, machine learning, deep learning, and generative models, and select the appropriate model family for specific behavioral prediction tasks
• Evaluate segmentation strategies beyond demographics and life stages and build intent-prediction approaches to map customer journeys from behavioral signals