Waze is a free navigation app that makes it easier for drivers around the world to get to where they want to go. Waze leadership wants to optimize the company’s retention strategy, enhance user experience, and make data-driven decisions about product development. They would like an analysis of WAZE data to understand their users better and the development of a machine learning model that predicts user churn. (Churn is understood to be the number of users who have uninstalled the Waze app or stopped using it.) This project is part of a larger effort at Waze to increase growth. It assumes that high retention rates indicate satisfied users who repeatedly employ the Waze app over time. Identifying and predicting which users are likely to churn will allow the WAZE team to target such individuals to induce their retention, thereby allowing Waze to grow its business.
Data
Waze’s free navigation app makes it easier for drivers around the world to get to where they want to go. Waze’s community of map editors, beta testers, translators, partners, and users help make each drive better and safer. Waze partners with cities, transportation authorities, broadcasters, businesses, and first responders to help as many people as possible travel more efficiently and safely. The data set is in-house from Waze for Cities (https://www.transportation.gov/office-policy/transportation-policy/faq-waze-data).
Deliverables
(Since this is an exercise, all models are predetermined. All Python code can be located at:
https://github.com/izsolnay/WAZE_Python)
- I. An analysis of WAZE data to understand their users better
- II. The development of a machine learning model that predicts user churn
- a. A binomial logistic regression model
- b. A winning tree-based model
- Appendix: A 2 sample t-test based on a sample of user data determining if there is a statistically significant difference in the mean number of rides between iPhone® users and Android™ users