ML_models/Project7

Overview The goal of this model is to predict whether or not a Waze user is retained or churned.

This modeling has three parts:

Part 1: Ethical considerations Should the objective of the model be adjusted?

Part 2: Feature engineering Perform feature selection, extraction, and transformation to prepare the data for modeling

Part 3: Modeling Build the models, evaluate them, and advise on the next steps.

Business Scenario

The Leadership at Waze (mobile app for drivers) has requested the development of a machine learning model to predict user churn in their app. To achieve the best results, it has been decided to build and test two tree-based models: Random Forest and XGBoost."

Data Understanding This project uses a dataset called waze_dataset.csv. It contains synthetic data created for this project in partnership with Waze. The data consisted of approximately 15k unique trips and 13 features. The features included information on: