📉 Customer Churn Prediction Using Machine Learning

📌 Project Overview

Customer churn is a critical challenge for subscription-based businesses.

This project builds an end-to-end machine learning pipeline to predict customer churn and identify key factors contributing to customer attrition.

The goal is to help businesses proactively retain customers by using data-driven insights.

This project uses separate training and testing datasets to ensure proper model evaluation.

🎯 Objectives

• Analyze customer behavior and churn patterns
• Perform exploratory data analysis (EDA)
• Engineer and preprocess features
• Build and compare machine learning models
• Evaluate models using appropriate metrics
• Derive actionable business insights

📊 Dataset

• Two datasets are used in this project:
  • customer_churn_dataset-training-master.csv
  • customer_churn_dataset-testing-master.csv
• The datasets contain customer demographic, subscription, and usage information
• Target variable: Churn (1 = Churned, 0 = Retained)
• Includes both numerical and categorical features

🛠️ Technologies & Libraries

• Python - Programming Language
• Pandas & NumPy – Data manipulation
• Matplotlib & Seaborn – Data visualization
• Scikit-learn – Machine learning models & evaluation

🔍 Exploratory Data Analysis (EDA)

Key analyses performed:

• Churn distribution analysis
• Churn rate by contract length
• Feature correlation analysis
• Identification of high-risk customer segments

⚙️ Data Preprocessing

• Converted categorical variables into numerical format using one-hot encoding
• Standardized numerical features to ensure consistent model performance
• Split data into training, validation, and testing sets for reliable evaluation
• Ensured feature consistency between training and testing datasets after encoding

🤖 Models Implemented

• Logistic Regression
• Random Forest Classifier

Both models were evaluated and compared using:

• Accuracy
• Precision
• Recall
• ROC-AUC Score
• Confusion Matrix
• ROC Curve Visualization

📈 Model Evaluation

• ROC curves were plotted for model comparison
• Random Forest achieved a strong predictive performance
• Feature importance analysis highlighted key churn drivers

💡 Key Business Insights

• Customers with short-term or month-to-month contracts show higher churn rates
• Certain subscription types are more prone to churn
• Contract duration and service usage play a major role in retention

🧠 Conclusion

This project demonstrates how machine learning can be applied to:

• Predict customer churn effectively
• Identify high-risk customers
• Support business decisions with actionable insights

▶️ How to Run the Project

1. Clone the repository

2. Install required libraries:

   pip install pandas numpy matplotlib seaborn scikit-learn
   

3. Open the notebook:

   Customer Churn Prediction.ipynb
   

4. Ensure the following files are in the same directory:

   • customer_churn_dataset-training-master.csv
   • customer_churn_dataset-testing-master.csv

5. Run all cells sequentially

👤 Author

Soundharya Sundaram Iyer

Aspiring Data Scientist / Passionate about Machine Learning

📌 Feel free to connect with me on LinkedIn and explore more projects on GitHub.

LinkedIn : https://www.linkedin.com/in/soundharyaiyer09

Github : https://github.com/Soundharya09

If you like this project, feel free to ⭐ star it!!!!!