business analytics course homework assignments

Churn Modelling using XGBoost

Behavioral segmentation and churn analysis (POV: Product Analytics Lead)

Part 3 of the D2C churn project: predictive churn modeling using machine learning. Trains and evaluates classification models to estimate churn probability, with feature engineering, metrics tracking, and explainability for business decisions.

I analyzed a dataset of telco customers' activity and how it relates to churn. This repository is designed for data cleaning and exploratory data analysis (EDA), which will be expanded into feature engineering and modelling.

Revenue concentration intelligence and predictive churn modeling framework with quantified financial exposure impact.

Sim2real comparison of user engagement and churn modeling using simulated anime viewing data and real Steam gameplay data, highlighting how data constraints shape modeling decisions.