This project analyzes customer churn within the telecommunications industry using the IBM Telco dataset. We implemented a dual-modeling approach: 1) A classification analysis using Logistic Regression and Random Forest to predict at-risk customers (achieving an AUC of 0.84), and 2) A linear regression audit to deconstruct the pricing drivers of monthly charges (R² ≈ 0.999). The findings provide actionable insights for targeted customer retention and pricing transparency.

This project analyzes customer churn within the telecommunications industry using the IBM Telco dataset. We implemented a dual-modeling approach: 1) A classification analysis using Logistic Regression and Random Forest to predict at-risk customers (achieving an AUC of 0.84), and 2) A linear regression audit to deconstruct the pricing drivers of monthly charges (R² ≈ 0.999). The findings provide actionable insights for targeted customer retention and pricing transparency.