About

This canvas implements a comprehensive end-to-end churn prediction system for telecom customers using Logistic Regression (80.2% accuracy, 84.2% ROC-AUC), featuring data pipeline automation, rigorous quality audits, risk segmentation, and a production-ready API with actionable retention strategies. The workflow loads customer data, cleans and encodes 19 features, trains and validates the model with extensive overfitting/leakage checks, then segments customers into risk tiers and deploys predictions through an API with business recommendations for proactive retention targeting.