Customer Churn Prediction

📌 Overview

This project analyzes customer behavior and builds a machine learning model to predict customer churn.

🎯 Objective

To identify customers likely to churn and understand the key factors influencing churn.

🛠️ Tools & Technologies

• Python (Pandas, NumPy)
• Data Visualization (Matplotlib, Seaborn)
• Machine Learning (Scikit-learn)
• Jupyter Notebook

📊 Key Steps

1. Data Cleaning and Preprocessing
2. Exploratory Data Analysis (EDA)
3. Feature Engineering
4. Model Building (Logistic Regression)
5. Model Evaluation

📈 Results

• Model Accuracy: ~79%
• Identified key churn factors:
  • Tenure
  • Monthly Charges
  • Contract Type
  • Payment Method

🧠 Key Insights

• Customers with shorter tenure are more likely to churn
• Higher monthly charges correlate with increased churn
• Contract type significantly influences retention

🚀 Future Improvements

• Handle class imbalance more effectively
• Try advanced models (Random Forest, XGBoost)
• Feature engineering for improved performance

📁 Project Structure

customer-churn-prediction/
├── data/
├── notebooks/
├── src/
└── requirements.txt

👤 Author

Hamzat Afe Isede

🔗 Related Projects

• Sales Forecasting (Time Series)
• Customer Segmentation
• Sales Dashboard (Power BI)