Customer Churn Prediction

Project Overview

This project predicts whether a telecom customer is likely to leave (churn) a service using Machine Learning. It includes data preprocessing, exploratory data analysis (EDA), data visualization, model training, and performance evaluation using a Random Forest Classifier.

---

Technologies Used

• Python
• Pandas
• NumPy
• Matplotlib
• Seaborn
• Scikit-learn

---

Dataset

• Dataset: Telco Customer Churn Dataset
• Total Records: 7,043
• Features: 21

---

Machine Learning Model

• Random Forest Classifier

---

Results

Metric	Value
Accuracy	78.92%
ROC-AUC Score	0.6834

---

Project Visualizations

1. Customer Churn Distribution

---

2. Monthly Charges vs Churn

---

3. Contract Type vs Churn

---

4. Confusion Matrix

---

5. Feature Importance

---

Project Structure

Customer_Churn_Prediction/
│
├── data/
│   └── telco_churn.csv
│
├── images/
│   ├── 01_churn_distribution.png
│   ├── 02_monthly_charges_vs_churn.png
│   ├── 03_contract_vs_churn.png
│   ├── 04_confusion_matrix.png
│   └── 05_feature_importance.png
│
├── customer.py
├── README.md
├── requirements.txt
└── .gitignore

---

How to Run

1. Clone the repository

git clone https://github.com/durgamsneha99/Customer-_Churn_Prediction.git

2. Open the project

cd Customer-_Churn_Prediction

3. Install dependencies

pip install -r requirements.txt

4. Run the project

python customer.py

---

Future Improvements

• Improve model accuracy using XGBoost or LightGBM.
• Perform hyperparameter tuning.
• Build a web application using Flask or Streamlit.
• Deploy the model for real-time predictions.

---

Author

Sneha Durgam

GitHub: https://github.com/durgamsneha99