Pragmatic Applications Series: Improving Churn Models

You are a machine learning engineer at a credit card company who is responsible for building a model to predict customer churn. After building your model, you will need to monitor its performance, drift, as well as any data quality issues in production. Arize will show you how to monitor and troubleshoot performance, drift and data quality issues in production.

In this workshop, you’ll learn best practices for how to:

• Set-up performance, drift and data quality monitoring to better understand how your model is performing. 
• Discover feature drifts corresponding to time periods of performance degradation and how to resolve them.
• Check to see if explainability and algorithm bias are having an impact on your model decisions.