Purchase Intent Prediction App

This project predicts whether a user will make a purchase based on their browsing behavior. It uses ensemble learning by combining LightGBM and CatBoost models for improved accuracy and recall. The Streamlit web app allows users to input behavioral data such as pages visited, duration, bounce rates, and session details to get real-time predictions on purchase intent.

Tech stack

• Python
• Streamlit
• LightGBM
• CatBoost
• Scikit-learn
• Joblib