Bayesian Networks

About the course

Our one-day training will start with an outline of how graph theory can be used to visualize classification problems and deepen your insights. We then move onto an in-depth discussion of the Bayes theorem, explaining the conditional probabilities, chain rule, and practical uses, along with interpreting the results of the Bayesian network and using Naïve-Bayes for classification problems. Finally, we address advanced classification networks using Tree Augmented Networks and Targeted Bayesian Network Learning. Challenge yourself with Bayesian Networks and you will walk away with these skills to successfully incorporate graph models. Rocketing your success and ability to understand and explain both your model output and variable interactions.  

Why this is for you

Understanding the main drivers of your model and their interactions are key to successful modeling. Bayesian belief nets are one of the most important probabilistic modeling methods in data mining and machine learning, with unique properties that are not present in other models. The Bayesian theory supports diagnosis rather than just prediction, thus it is explanatory, helping the user understand some counter-intuitive phenomena and uncover the causal mechanisms that affect the target variable.  

For whom

This training is designed for Data Scientists who have completed our Machine Learning Process (3201) badge. This course deals with complex statistical and modeling concepts, therefore, participants must have a comprehensive understanding of mathematics and existing knowledge of model theory and programming language R or Python to succeed.  

What you’ll learn

This training will work with the concept of variable importance within different modeling settings, SHAP-values, Bayes’ theorem, Tree Augmented Networks (TAN) theory, Targeted Bayesian Network Learning (TBNL) theory, and extended classification and model interpretation cases in R or Python. You will learn:

1. How graph theory works
2. The conditional probabilities and chain rules of the Bayes theorem
3. How to interpret a Bayesian network
4. The Naïve-Bayes model
5. Advanced classification networks TAN and TBNL

Learning Goals

• Introduction to graph theory – Understand and explain how graph theory can be used to visualize classification problems
• Fundamentals of Bayes’ theorem – Know the conditional probabilities and chain rules of Bayes’ theorem
• Understanding Bayesian Networks – Capable of interpreting the results of a Bayesian network
• Basic classification networks – Perform classification and interpret results using Naïve-Bayes
• Advanced classification networks – Use TAN and TBNL for classification problems

Theory and practical use All trainings in the GAIn portfolio combine high-quality standardized training material with theory sessions from experts and hands-on experience where you directly apply the material to real-life cases. Each training is developed by top of the field practitioners which means they are full of industry examples along with practical challenges and know-how, fueling the interactive discussions during training. We believe this multi-level approach creates the ideal learning environment for participants to thrive.