Data Modelling for AI-Driven BI

This session explores how to design data models that support machine learning within BI environments. It covers techniques for feature engineering, temporal modelling, and integrating unstructured data. It’s ideal for teams aiming to operationalize AI in scalable, real-time BI workflows.

Key Takeaways:
1. Design feature-ready data models that support robust ML workflows and business logic.
2. Apply temporal modelling to capture time-based patterns and improve predictive accuracy.
3. Integrate unstructured data (e.g., text, images) into structured BI environments.
4. Build semantic layers for consistent, explainable, and governed analytics.
5. Avoid common modelling pitfalls like data leakage, poor granularity, and overfitting.

- Practical techniques to design feature ready data models that bridge BI and ML - Clear understanding of temporal modelling, unstructured data integration, and avoiding data leakage - Actionable guidance to build explainable, governed semantic layers that scale AI in real world BI workflows.