How vemulasailahith/customer_churn_prediction_system works
This is a small Streamlit customer-churn application. A user enters customer profile, service, plan, financial, and value information, then requests a prediction. The app preprocesses the input, obtains churn classification and probability, and scales segmentation features before assigning a K-Means cluster and displaying the results. README capabilities and the sampled app support this flow; segmentation implementation details are partly omitted, so the graph shows only the documented scaler-to-cluster relationship.
Main entry points: app.py Streamlit App · customer_churn_artifacts.pkl Churn Model
Components 9
Application
- Streamlit App
app.py - Customer Inputs
app.py - Prediction Results
app.py
Model Artifacts
- Preprocessor
customer_churn_artifacts.pkl - Churn Model
customer_churn_artifacts.pkl - Feature Scaler
customer_churn_artifacts.pkl - K-Means Model
customer_churn_artifacts.pkl - Trained Artifacts
customer_churn_artifacts.pkl
Other components
- Customer
Connections 12
- Customer → Customer Inputs: enters details
- Customer Inputs → Streamlit App: submits inputs
- Streamlit App → Trained Artifacts: loads
- Streamlit App → Preprocessor: transforms inputs
- Preprocessor → Streamlit App: returns features
- Streamlit App → Churn Model: predicts churn
- Churn Model → Streamlit App: returns prediction
- Churn Model → Streamlit App: returns probability
- Streamlit App → Feature Scaler: scales features
- Streamlit App → K-Means Model: assigns cluster
- Streamlit App → Prediction Results: displays results
- Prediction Results → Customer: shows output