Customer Churn Prediction Project

📌 Project Overview

This project predicts customer churn using machine learning. The goal is to analyze customer behavior and identify which users are likely to leave a service. Businesses can use this model to take proactive actions like retention offers and better support.

🧠 Problem Solved

Companies lose revenue when customers cancel subscriptions.
This project helps identify customers at high risk of churn so the business can:

• Improve customer retention strategy
• Give personalized offers to risky customers
• Reduce overall churn rate
• Save marketing & re-acquisition cost

👤 Author

Yuvraj
Aspiring ML engineer from India
📧 Email: y.india.main@gmail.com

✅ Steps Followed

1. Problem Understanding

• Business problem: Identify customers likely to churn
• Type: Binary classification (0 = stays, 1 = churn)

2. Data Loading & Exploration

• Loaded dataset
• Checked missing values & data types
• EDA: distributions, churn ratio, correlations

3. Data Preprocessing

• Missing value handling
• Categorical encoding
• Feature/target split
• Train-test split (80-20)
• StandardScaler normalization for numeric features

4. Model Training

Trained models:

• Logistic Regression
• Random Forest Classifier ✅ selected

Metrics used:

• Accuracy
• ROC-AUC
• Confusion Matrix
• Precision/Recall/F1-score

5. Model Saving

Files saved in models/

File	Description
churn_model.pkl	Trained model
scaler.pkl	StandardScaler
model_features.pkl	Feature list

Saved using joblib.dump(...).

6. Inference Pipeline

Created predict.py to:

• Load model + scaler + features
• Preprocess new data
• Predict churn + probability

7. Deployment Plan

• Streamlit interface
• Input form + results + charts
• Host on Streamlit Cloud / Render

📂 Project File Structure

project/
 ├── data/                  
 ├── notebooks/             
 ├── models/                
 ├── predict.py             
 ├── app/streamlit_app.py   
 ├── requirements.txt
 └── README.md

🛠️ Skills Used

• Python
• Pandas, NumPy
• Matplotlib, Seaborn
• Scikit-Learn
• Machine Learning (Classification)
• Model Serialization (joblib)
• Streamlit (deployment)

🎯 Deliverables

• EDA + model training notebook
• Saved model and scaler
• Prediction script
• Streamlit web app

---

Thanks for reading!
Have a productive day and keep learning 🔥