Applied ML/Team · Applied machine learning
Customer Churn Prediction.
Turning customer data into better questions.
Illustrative data visualization. The exact individual contribution, dataset, and evaluation methodology are being documented.
Understanding the problem.
Customer retention depends on understanding behavioral signals. Historical customer data can help structure the question of who may leave and why.
Connecting the pieces.
A team project following CRISP-DM, using SQL to process historical customer data and five behavioral variables for customer-loyalty and churn prediction.
Where I contributed.
- Participated in a team-based data mining project.
- Applied a structured business-understanding-to-modeling workflow.
- Connected prediction outputs to customer-retention analysis.
Results, with context.
The source brief reports 85% model accuracy. Dataset details, the algorithm, train/test methodology, and metrics beyond accuracy remain to be validated; this figure alone does not establish model performance.
UP NEXT