Causal Inference - How We Are Applying Causal Methods in Retail Data Science, Key Tips and Learnings

Learn how to apply causal inference methods in retail data science through practical insights from dunnhumby's Senior Research Data Scientist Dr Dimitra (Mimie) Liotsiou in this 20-minute conference talk. Discover why standard data science and machine learning methods focus on correlation rather than causation, and understand the limitations this creates when trying to answer cause-and-effect questions about measuring impacts, uplift, or KPI movements that are central to many data science projects. Explore the new science of graphical causal inference and see real-world applications of these methods in retail environments. Gain valuable tips and learnings from practical implementation of causal methods, moving beyond traditional correlation-based approaches to develop a more comprehensive understanding of how to answer critical business questions about causation in data science projects across various sectors.