Overview
Customer Churn Prediction is a machine learning project aimed at predicting customer attrition for a telecommunications company. The system utilizes historical customer data to train a predictive model that identifies patterns associated with customer churn.
Performed data preprocessing, feature engineering, and trained multiple ML models including Logistic Regression, Random Forest, and XGBoost. Built a REST API using FastAPI to serve the trained model. Designed an interactive web interface using Next.js for real-time predictions.
Performed data preprocessing, feature engineering, and trained multiple ML models including Logistic Regression, Random Forest, and XGBoost. Built a REST API using FastAPI to serve the trained model. Designed an interactive web interface using Next.js for real-time predictions.
Technologies
- Next.js 15
- NumPy
- FastAPI
- Jupyter
- Scikit-learn
# Features
- Worked on Telco Customer Churn dataset (7043 records, 30 features)
- Performed data cleaning, encoding, and feature engineering for model readiness
- Trained and compared multiple models (Logistic Regression, Random Forest, XGBoost)
- Achieved best performance with Logistic Regression: Accuracy: 82%, F1 Score: 0.64, ROC-AUC: 0.74