Projects / Computer Science / Computer Science project 5

Customer Churn Prediction with a Business Dashboard

Approx. component budget: Rs. 0 - 1,000

Objective

Predict churn from usage and billing data and turn the result into actions a business team could really take.

What you will build

A model that predicts which telecom or subscription customers may leave, plus a dashboard showing risk groups and the main reasons behind churn.

Results you can expect

• ROC-AUC and recall for the churn class
• Handling of class imbalance with SMOTE or class weights
• Top 5 churn drivers with SHAP
• And a lift chart for the top 20 percent risky customers

What you will learn

Feature engineering, imbalanced data, model evaluation beyond accuracy, and presenting data to non-technical people.

Tools and notes

Python, pandas, scikit-learn, XGBoost, SHAP, Power BI or Streamlit, public telecom churn dataset.

Resume points from this project

• Built and tested Customer Churn Prediction with a Business Dashboard; objective: predict churn from usage and billing data and turn the result into actions a business team could really take.
• Measured and analysed rOC-AUC and recall for the churn class and handling of class imbalance with SMOTE or class weights; wrote the report and presented the results in a viva. Add your own numbers: [x].

Replace [x] with your own measured results. Write only what you really did.

More computer science projects

Back to all projects