This project aims to address user churn for the Waze app by developing a machine learning model that predicts the likelihood of users discontinuing app use. By analyzing user behavior data, this project offers insights and actionable strategies to help Waze improve user retention and target high-risk users effectively.
The primary objective of this project was to build a predictive model to identify users at risk of churn. The model analyzes user behavior metrics such as session frequency and kilometers driven, providing valuable insights to enhance retention strategies.
- Model Performance: Achieved an 87% accuracy score using the XGBoost algorithm.
- Key Insights: The analysis showed that users with high session frequency and substantial kilometers driven were at a higher risk of churning.
This project led to several actionable recommendations for Waze:
- Retention Strategies: By identifying key indicators of churn, Waze can focus on targeting high-risk users (e.g., long-distance drivers) with personalized strategies to retain them.
- Potential Reduction in Churn: Targeted retention efforts based on these findings could help reduce churn rates by focusing on users with specific behavioral patterns.
- Programming Language: Python
- Libraries:
- Data Analysis:
Pandas,NumPy - Machine Learning:
Scikit-learn,XGBoost - Data Visualization:
Matplotlib,Seaborn
- Data Analysis:
-
Clone the Repository:
git clone https://github.com/your-username/waze-user-churn-prediction.git cd waze-user-churn-prediction -
Install Dependencies: Ensure you have Python installed:
-
Run the Model: Follow the instructions in the notebook, train the model, and evaluate performance.
The model predicts a user's churn probability based on their usage patterns.
This project is licensed under the MIT License.
This project was developed as part of the Google Advanced Data Analysis Course on Coursera. It is intended solely for educational and portfolio purposes.