Skip to main content
Workshop exercise
In Progress

Model evaluation practical

In this exercise, you will develop and evaluate Random Forest, Gradient Boosting, and XGBoost models using demographic and behavioural customer data to predict churn.

Instruction icon

Challenge instructions and resources

Download the Jupyter notebook bundle for the practical activity.

Business context​

  • ** Business:** RingNet, a UK telecommunications provider.
  • ** Problem:** Rising customer churn in a competitive market.
  • ** Task:** Build and evaluate ensemble models that identify customers likely to churn.
  • ** Role:** You are a Machine Learning Engineer in the Data & Insights team.

Activity flow​

  • Work through the notebook instructions.
  • Compare model performance and refine where needed.
  • Discuss approaches in the breakout room.
  • Summarise your findings before regrouping after 25 minutes.