Predicting the Customer Behavior Utilizing Tree Based Machine Learning Algorithms

This project examines a tree-based machine learning approach to predict customer behavior outcomes in ecommerce, using a large dataset. The project compares different classification methods to solve three Customer Relationships Management problems: predicting customer satisfaction, churn modeling, and the next product to buy modeling. The analysis is fully automated, making it easy for small e-retailers to implement. The study employs decision tree, random forest, and gradient boosting techniques.