In the workshop, we will explore how to develop an interpretable machine learning classification model in Python. We will focus on using decision trees, which allow for intuitive interpretation of the results. As a practical example, we will use the well-known Titanic passenger dataset. Participants will develop and visualize the model step by step.
Content
Setting up the Python environment and Jupyter Notebook.
Introduction to machine learning and model interpretability.
Analysis of the training dataset (Titanic).
Building and interpreting a decision tree.
Evaluating model performance and visualizing results.
Recommended prior knowledge
Basic knowledge of programming concepts is recommended. No prior experience in machine learning is required.
Learning objectives
Understand the basics of machine learning and the importance of model explainability
Independently develop decision trees using Python
Visualize rules and feature importance
Evaluate model performance using basic metrics
Target group
Data analysts and developers who want to understand how predictive models work and how to interpret them.