Machine Learning

Customer Churn Prediction

This project predicts customer churn in a telecom company using XGBoost. The model identifies patterns in customer behavior, contract type, and service usage to predict which customers are likely to leave.

Model Accuracy

91%

Model Type

XGBoost

Category

Machine Learning

Tech Stack

4 Tools

Key Features

  • Class imbalance handling
  • Feature scaling and normalization
  • XGBoost classification
  • Feature importance visualization
  • Business impact analysis

Technologies

PythonXGBoostScikit-LearnPandas

Performance Metrics

Accuracy

91

Precision

89

Recall

92

AUC-ROC

94

Feature Importance

Churn Distribution

Interested in this project?