Infinite Churn Analysis

Intro

We have built an interpretable ML-driven Churn predictor.

Challenges faced

We faced three major challenges while working on this project:

1. Handling missing data
2. Handling class imbalance
3. Identifying relevant features for prediction

Dealing with these challenges made us realise that real-world data is often messy and imperfect.

ML MODEL

We used a LightGBM model which allows us to make decisions about customers.

Built With

• pandas
• python